Cancel job
Stop a job started by a long call, and ask the work behind it to stop too. Use it when the user no longer wants the answer; a job left alone keeps running and can be read later with check_job.
ActionTry it
Integrate Keiki MCP
with your AI CRM
Build and manage Keiki agents, conversations, evaluations, channels, billing, integrations, and organization settings through the official hosted MCP server.
Explore Triggers and Actions
Check job
Read a job that a long call handed back instead of a result: its status, the text and tool calls it has produced so far, and its final result once it has finished. Pass waitMs to have this call wait for the job to finish (up to 30000ms) rather than answering immediately.
ActionTry it
Current context
Who this token acts as, which organization it is bound to, and the current time.
ActionTry it
Delete agent groups by id
Delete an agent group, so its members can no longer ask each other through it. Owners and admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /agent-groups/:id → { status, ok, data, error }
ActionTry it
Delete agents by id
Delete an agent with its tools, alerts, loops and configuration, releasing its Keiki VM if it has one. SDK-managed (local) agents are refused. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /agents/:id → { status, ok, data, error }
ActionTry it
Delete agents capabilities by slug
Uninstall a capability from an agent, dropping its version pin. Owners and admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /agents/:id/capabilities/:publisherRef/:slug → { status, ok, data, error }
ActionTry it
Delete agents identity webhook
Remove sender identity checks; returns { ok }. Owners/admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /agents/:id/identity-webhook → { status, ok, data, error }
ActionTry it
Delete agents lines by number
Remove a phone line from an agent; the agent stops receiving on it. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /agents/:id/lines/:number → { status, ok, data, error }
ActionTry it
Delete agents loops by loop id
Delete one of the agent’s scheduled loops for good: its schedule stops and its run history is removed. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /agents/:id/loops/:loopId → { status, ok, data, error }
ActionTry it
Delete agents mcp presets by preset id
Disconnect an MCP preset from the agent and remove it from its daemon. A connection shared by the organization cannot be removed here; change its grants instead. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /agents/:id/mcp-presets/:presetId → { status, ok, data, error }
ActionTry it
Delete agents tools by tool id
Delete an authored tool. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /agents/:id/tools/:toolId → { status, ok, data, error }
ActionTry it
Delete agents wallet intents by intent id
Revoke a purchase: cancels its payment credential so it can no longer be charged, and returns the updated purchase. Owners and admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /agents/:id/wallet/intents/:intentId → { status, ok, data, error }
ActionTry it
Delete conversations block
Unblock one identity for this conversation’s agent so its messages trigger replies again. Also how an ended inter-agent thread is reopened. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /conversations/:phone/block → { status, ok, data, error }
ActionTry it
Delete conversations contact card
Clear the dynamic contact card for this conversation. Clearing re-enables the native daily contact-card share. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /conversations/:phone/contact-card → { status, ok, data, error }
ActionTry it
Delete conversations history
Permanently delete one agent’s stored message history with an identity and return how many messages were cleared. The identity must be unblocked first (409 otherwise); opt-out status is unchanged. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /conversations/:phone/history → { status, ok, data, error }
ActionTry it
Delete conversations takeover
End a human takeover early and hand the conversation back to the agent, including any hold taken from a Slack thread. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /conversations/:phone/takeover → { status, ok, data, error }
ActionTry it
Delete deployment
Delete one deployment and release its connected channels.
ActionTry it
Delete evals cases by id
Delete an organization eval case permanently, not just archive it. Returns ok even if the case is absent. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /evals/cases/:id → { status, ok, data, error }
ActionTry it
Delete identities by id
Delete an identity: tears down its inbound routing and deactivates it, so the channel stops answering. The provider-side app or number is untouched. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /identities/:id → { status, ok, data, error }
ActionTry it
Delete identities slack installs by install id
Stop answering in one workspace that installed this Slack identity, without touching the app or its other workspaces — the workspace keeps the app in Slack; only its stored token is dropped. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /identities/:id/slack/installs/:installId → { status, ok, data, error }
ActionTry it
Delete identities slack operators by slack user id
Remove one Slack user from a Slack identity’s operator allowlist. Owners and admins only. Returns the resulting list. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /identities/:id/slack/operators/:slackUserId → { status, ok, data, error }
ActionTry it
Delete identities slack teams by team id
Stop routing one customer Slack workspace to its agent. The install stays; the workspace gets an “unbound” notice until it is bound again. Owners and admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /identities/:id/slack/teams/:teamId → { status, ok, data, error }
ActionTry it
Delete identities teams installs by install id
Forget one tenant’s install of a Teams identity so the agent stops answering there. Teams still has the app installed; the next verified activity from that tenant would recreate the install. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /identities/:id/teams/installs/:installId → { status, ok, data, error }
ActionTry it
Delete invitations by id
Revoke a pending invitation so its link no longer works. Owners and admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /invitations/:id → { status, ok, data, error }
ActionTry it
Delete mcp clients by token id
Revoke this user’s organization OAuth connection; returns { ok }. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /mcp-clients/:tokenId → { status, ok, data, error }
ActionTry it
Delete members by id
Remove a member from the organization. Owners and admins only; the owner cannot be removed (transfer ownership first) and you cannot remove yourself. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /members/:id → { status, ok, data, error }
ActionTry it
Delete org mcp integrations by id
Stop sharing an org MCP integration, removing it from every agent it was granted to. Owners and admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /org-mcp-integrations/:id → { status, ok, data, error }
ActionTry it
Delete payments
Disconnect payments; returns { ok }. Owners/admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /payments → { status, ok, data, error }
ActionTry it
Delete payments payees by id
Delete a payee; returns { ok }. Owners/admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /payments/payees/:id → { status, ok, data, error }
ActionTry it
Delete provider accounts by id
Disconnect one of the organization’s own Photon accounts. Owners and admins only. Removes Keiki’s webhook from the Photon project, forgets its stored secrets, takes its lines offline and unassigns them from agents. The lines stay in the customer’s Photon project. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /provider-accounts/:id → { status, ok, data, error }
ActionTry it
Delete sandbox scripts by id
Delete a sandbox script; agents it is attached to stop running it. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /sandbox-scripts/:id → { status, ok, data, error }
ActionTry it
Delete secrets by name
Delete a secret and its bound mirrors; returns { ok, name }. Configurers only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /secrets/:name → { status, ok, data, error }
ActionTry it
Delete skills by id
Delete a skill or document; agents using it lose it. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /skills/:id → { status, ok, data, error }
ActionTry it
Delete sso
Disable SSO; returns { ok }. Owners/admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /sso → { status, ok, data, error }
ActionTry it
Delete sso domains by id
Remove an SSO domain; returns { ok }. Eligible owners/admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /sso/domains/:id → { status, ok, data, error }
ActionTry it
Delete trace exporters by id
Delete a destination and its agent attachments. Owners, admins and members only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /trace-exporters/:id → { status, ok, data, error }
ActionTry it
Delete webhooks by id
Delete a webhook endpoint. 409 while an agent’s escalation route still uses it. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. DELETE /webhooks/:id → { status, ok, data, error }
ActionTry it
Deploy agent
Create or update an agent deployment using the body format of POST /api/v1/developer/deployments.
ActionTry it
Escalate
File a bug report or feature request about the Keiki platform itself with the engineering team. Use it when something is broken or missing in the platform — not for a problem the user can fix in their own agent's configuration, which you should fix or recommend instead. File once per finding, with the evidence you actually observed (trace ids, error strings, the endpoint and response, what the user reported); a trace is not required. Escalating fixes nothing: still tell the user what you found.
ActionTry it
Get agent goals
The outcome kinds each agent has produced and whether any carry money — what an agent’s CRM columns stand for. GET /agent-goals → { status, ok, data, error }
ActionTry it
Get agent groups
List the organization’s agent groups — which agents may ask each other with ask_agent — with each group’s name, description and members (agent, description, hub or spoke role). GET /agent-groups → { status, ok, data, error }
ActionTry it
Get agent groups activity
Every agent group’s running peer threads and live roster in one read — the way to answer how all the groups (swarms) are doing at once. GET /agent-groups/activity → { status, ok, data, error }
ActionTry it
Get agent groups note
The group’s shared note — what its agents have settled among themselves (decisions, who does what, findings) — with its revision, last author and update time. GET /agent-groups/:id/note → { status, ok, data, error }
ActionTry it
Get agent groups note versions
History of the group note, newest first: one entry per saved version, with the sha to fetch it by. GET /agent-groups/:id/note/versions → { status, ok, data, error }
ActionTry it
Get agent groups note versions by sha
The group note body as it was at one historical version (a sha from the versions list). GET /agent-groups/:id/note/versions/:sha → { status, ok, data, error }
ActionTry it
Get agent groups roster
Live roster of one group: per member its status (running, waiting, idle, ended), running turns, pending asks, background tasks and 24h tokens and cost, plus totals. The place to answer how a swarm is doing. GET /agent-groups/:id/roster → { status, ok, data, error }
ActionTry it
Get agent groups threads
The asker→peer threads currently live among one group’s members (who is asking whom right now). Empty for an unknown group. GET /agent-groups/:id/threads → { status, ok, data, error }
ActionTry it
Get agent templates
Complete agent templates the platform can provision in one call (e.g. Orchid): id, name, description, model, required secrets, user connections and pinned capabilities. Create one by passing its id as `template` when creating an agent. GET /agent-templates → { status, ok, data, error }
ActionTry it
Get agentmail domains
Custom email domains the organization added for agent inboxes, each with its region and live DNS verification state from AgentMail. GET /agentmail/domains → { status, ok, data, error }
ActionTry it
Get agents
List every agent in the organization with its status, model and channels. GET /agents → { status, ok, data, error }
ActionTry it
Get agents a2a
Whether other vendors’ agents can call the agent over A2A, with its Agent Card URL and JSON-RPC endpoint. GET /agents/:id/a2a → { status, ok, data, error }
ActionTry it
Get agents actions
List the agent’s persisted approval actions (tool calls waiting on, or past, a human decision), newest first, with their status, arguments and group id. GET /agents/:id/actions → { status, ok, data, error }
ActionTry it
Get agents analytics
Engagement for one agent over a date range: contact and agent response times, received/sent message breakdowns such as where contacts write from (null when event analytics is not configured), and public profile visits. GET /agents/:id/analytics → { status, ok, data, error }
ActionTry it
Get agents api gateway
Whether the agent answers on the OpenAI/Anthropic-compatible API and in passthrough mode, with the base URLs and the model name a client sends. GET /agents/:id/api-gateway → { status, ok, data, error }
ActionTry it
Get agents by id
Agent detail: its configuration and features, line routing status, harness, template version, tools, attached skills, its 50 most recent conversations, and connection issues to fix. GET /agents/:id → { status, ok, data, error }
ActionTry it
Get agents capabilities
List the marketplace capabilities installed on an agent, including disabled ones, with each pinned version, how it was installed, any newer version available, and declared secrets the org has not set yet. GET /agents/:id/capabilities → { status, ok, data, error }
ActionTry it
Get agents conversations
Every conversation one agent is having, newest first — who it is talking to, the last message, message count, whether the thread is active and whether it has errors. Start here for how an agent’s conversations are going, then open the ones that matter with /agents/:id/conversations/:phone. For money owed, pair it with /outcomes?status=pending&agent_id=… (a pending outcome committed and has not paid). GET /agents/:id/conversations → { status, ok, data, error }
ActionTry it
Get agents conversations by phone
One conversation thread between the agent and a contact (`phone` is their phone number, email or tg: identity, URL-encoded): its metadata, messages and the trace spans of the turns that produced them. GET /agents/:id/conversations/:phone → { status, ok, data, error }
ActionTry it
Get agents conversations sentiment
One conversation’s mood per user turn, oldest first (0 angry to 4 delighted; unclassified turns have no score), with drop/recovery shifts, the reply that caused each shift and why, and the signals on each turn. GET /agents/:id/conversations/:phone/sentiment → { status, ok, data, error }
ActionTry it
Get agents edit
The agent form payload: the full editable config (name, model, systemPrompt, features, escalationRoutes, skillIds, sandboxScriptIds, sandboxEnvSecrets, execPolicy, userSecrets, responseSchemaOverlay, updatedAt) plus the org’s lines. Always read it before updating. When `passthroughPrompt` is set, or the gateway is passthrough with it empty, the caller’s instructions lead each turn and `systemPrompt` is only an amendment to them; `responseSchemaOverlay` likewise only reorders and re-describes fields the caller’s response_format already declares. `responseSchemaLint` quotes prompt rules about field order the schema does not satisfy, and `responseSchemaLint.suggested` is the overlay to save as `responseSchemaOverlay` instead of rewording the prompt. GET /agents/:id/edit → { status, ok, data, error }
ActionTry it
Get agents evals channel preview
Preview a real-channel eval run per hosted driver (imessage, email, x): whether it is available and why not, the handle it would message, and the eligible and excluded cases — plus any active run and the last run. Needs a paid plan. GET /agents/:id/evals/channel/preview → { status, ok, data, error }
ActionTry it
Get agents executor
Resolve the agent’s Executor.sh sandbox daemon and return its current short-lived MCP URL, base URL and scope (never its auth token). Needs the agent’s Sandbox and MCP capabilities enabled; boots the daemon if needed. GET /agents/:id/executor → { status, ok, data, error }
ActionTry it
Get agents fine tune
Read fine-tune readiness for the agent’s current version: graded conversation counts, progress toward the kickoff threshold, whether the caller may start one, and any request already on file. GET /agents/:id/fine-tune → { status, ok, data, error }
ActionTry it
Get agents harness status
Report what runs the agent and whether it is connected: the platform harness, a self-hosted Hermes relay or dispatch URL (with why it is waiting), or a Maritime Keiki VM and its state. GET /agents/:id/harness-status → { status, ok, data, error }
ActionTry it
Get agents loops
List the scheduled loops attached to an agent with their schedule and state. GET /agents/:id/loops → { status, ok, data, error }
ActionTry it
Get agents loops runs
Read a loop’s recent runs, newest first, from a bounded retention window: status (the platform’s turn and tool-health view), timings, error, trace id and detail. A blocked_auth detail names the connection unavailable when the run started, not proof that it still is. GET /agents/:id/loops/:loopId/runs → { status, ok, data, error }
ActionTry it
Get agents mcp presets tools
The tools a connected MCP exposes, read live from the daemon, each with the permission (allow/ask/deny) in force, plus the default permission and every stored per-tool choice. GET /agents/:id/mcp-presets/:presetId/tools → { status, ok, data, error }
ActionTry it
Get agents note
Read the agent’s working note — the durable scratchpad it rewrites while it works — with its revision. GET /agents/:id/note → { status, ok, data, error }
ActionTry it
Get agents note versions
List the working note’s version history, newest first, with the sha of each version. GET /agents/:id/note/versions → { status, ok, data, error }
ActionTry it
Get agents note versions by sha
Read the working note body as it was at one version sha from the history. GET /agents/:id/note/versions/:sha → { status, ok, data, error }
ActionTry it
Get agents public profile
Whether the agent’s public profile page is published, its URL slug and tagline, and whether the caller may change it. GET /agents/:id/public-profile → { status, ok, data, error }
ActionTry it
Get agents quality
Turn-quality review for one agent: review stats, its review settings, and the prompt or model fixes the reviewer drafted (state, summary, rationale, proposed prompt/model, score before and after). GET /agents/:id/quality → { status, ok, data, error }
ActionTry it
Get agents rollouts
Read the agent’s open staged rollout (mode, percent of conversations on the staged config, the staged config itself) and its recent ended rollouts. GET /agents/:id/rollouts → { status, ok, data, error }
ActionTry it
Get agents sandbox script runs
The latest 50 runs of the agent’s sandbox scripts, newest first — status, exit code and output tails. Script failures degrade silently, so this is where a broken script shows up. GET /agents/:id/sandbox-script-runs → { status, ok, data, error }
ActionTry it
Get agents shared line
The shared iMessage number and join code contacts text to reach this agent. join is null when Keiki has no shared number, or without a plan (planRequired) unless ?test=1: onboarding's test, which keeps a plan-less agent answering on it for 15 minutes, capped like Live Test. GET /agents/:id/shared-line → { status, ok, data, error }
ActionTry it
Get agents signals
Built-in quality signals for one agent over a date range: each signal’s events, share of classified users and change against the previous period of the same length, plus average mood (0 angry to 4 delighted) and the biggest mood drops with the judge’s reason. GET /agents/:id/signals → { status, ok, data, error }
ActionTry it
Get agents signals events
The turns one signal (`key`, as returned by the signals summary) fired on, newest first: phone, trace id, source, mood and shift. Pass `nextCursor` from a page as `cursor` to read the next one. GET /agents/:id/signals/:key/events → { status, ok, data, error }
ActionTry it
Get agents user connections
List the personal provider definitions declared on the agent — services each conversation user authorizes with their own account. No credentials are returned. Read this before replacing them. GET /agents/:id/user-connections → { status, ok, data, error }
ActionTry it
Get agents versions
List the agent configuration’s version history, with the sha of each version. GET /agents/:id/versions → { status, ok, data, error }
ActionTry it
Get agents versions by sha
Read the agent configuration snapshot saved at one version sha. GET /agents/:id/versions/:sha → { status, ok, data, error }
ActionTry it
Get agents wallet
The agent’s wallet: spending policy, connection state, spend so far, pending purchase approvals and history — or null when it has none — plus the payment programme, default policy, and whether the caller may manage it. GET /agents/:id/wallet → { status, ok, data, error }
ActionTry it
Get agents wallet payment methods
Labelled payment methods on the wallet’s connected account (never card numbers), for choosing `paymentMethodId`. Owners and admins only. GET /agents/:id/wallet/payment-methods → { status, ok, data, error }
ActionTry it
Get audit
List audit events as { events, nextCursor }; filter by cursor, limit, actor, action, target_type, target_id, from, to. Owners/admins only. GET /audit → { status, ok, data, error }
ActionTry it
Get audit verify
Verify the organization audit hash chain and return its verification result. Owners/admins only. GET /audit/verify → { status, ok, data, error }
ActionTry it
Get auto
Read whether the organization’s plan allows Auto mode and which front and worker models it uses. GET /auto → { status, ok, data, error }
ActionTry it
Get axiom traces
Spans read from the OTLP collector — kept past the 14-day window of /traces, and alongside spans other services exported under the same trace. Returns { range, spans }. 503 when no collector is configured; use /traces then. GET /axiom/traces → { status, ok, data, error }
ActionTry it
Get billing
Read the organization’s billing overview: available phone lines, usage, customer plan details, upcoming invoice and default plan id. The upcoming invoice may be a mid-cycle estimate. GET /billing → { status, ok, data, error }
ActionTry it
Get billing activity
List the organization’s invoice and platform credit grant activity, newest first. Returns items and nextCursor; pass nextCursor as cursor to read the next page. GET /billing/activity → { status, ok, data, error }
ActionTry it
Get billing auto top up
Read the organization’s auto top-up settings: enabled, threshold, quantity, whether it is available on the plan, whether a payment method is saved, and the last failed top-up if any. GET /billing/auto-top-up → { status, ok, data, error }
ActionTry it
Get billing capability pricing
Read the default plan’s per-capability billing metadata, included amounts, units and price labels. Returns capabilities for explaining costs and BYOK offsets, not customer-specific usage. GET /billing/capability-pricing → { status, ok, data, error }
ActionTry it
Get billing capability usage by agent
Read local capability usage grouped by feature and agent for each feature’s latest recorded billing period, including unattributed and deleted-agent buckets. Returns features with units and period ends; this is platform attribution, not Autumn per-agent billing or invoice reconciliation, and capabilities share the ai_tokens meter. GET /billing/capability-usage-by-agent → { status, ok, data, error }
ActionTry it
Get billing consumer pricing
Read public consumer plans, add-ons and feature pricing metadata from the billing catalog. Returns display-only prices and allowances, or an unavailable error when pricing cannot be loaded. GET /billing/consumer-pricing → { status, ok, data, error }
ActionTry it
Get billing included allowance
Read public default-plan included token-credit allowances without customer data. Returns allowances with feature id, positive included amount and denomination, deduplicated by feature. GET /billing/included-allowance → { status, ok, data, error }
ActionTry it
Get billing plans
List the organization’s available non-add-on, non-archived plans with display prices, features, trial information and current/scheduled flags. Returns plans, or an empty list if the catalog cannot be loaded. GET /billing/plans → { status, ok, data, error }
ActionTry it
Get billing portal
Create a Stripe billing portal link for the organization. Returns url to share with the user; do not follow it yourself. GET /billing/portal → { status, ok, data, error }
ActionTry it
Get billing pricing account
Read the connected organization’s current plan and status, usage and credit access checks. Returns customer, usage, access and canManage; canManage indicates whether the caller is an owner or admin. GET /billing/pricing-account → { status, ok, data, error }
ActionTry it
Get billing public plans
Read public recurring plans with included limits and platform credit conversion rates for comparing options. Returns plans and rates without customer data, or an unavailable error. GET /billing/public-plans → { status, ok, data, error }
ActionTry it
Get billing service spend
Read the organization’s recorded customer service charges for the billing period. Returns itemized services, total, unassigned remainder and reconciliation status, or available false if no currency balance exists. GET /billing/service-spend → { status, ok, data, error }
ActionTry it
Get billing status
Check the organization’s lightweight plan gate rather than loading its billing overview. Returns sandbox, indicating plan-less or unavailable billing, and the default plan id. GET /billing/status → { status, ok, data, error }
ActionTry it
Get billing usage by conversation
List up to 100 conversations with the most recorded token usage across the organization’s active developer configurations, excluding playground and copilot traffic. Returns all-time turn, token, tool-call and duration totals, models and last-turn time per conversation, not monetary costs or a billing-period filter. GET /billing/usage-by-conversation → { status, ok, data, error }
ActionTry it
Get blog
Keiki's own blog, newest first: each post's slug, title, category (Changelog, Product, Announcements), publish date, summary and URL. Read one with GET /blog/:slug. GET /blog → { status, ok, data, error }
ActionTry it
Get blog by slug
One Keiki blog post in full: its title, category, date and summary, plus the whole post as `markdown`. Changelog posts list what shipped on Keiki and when. 404 for an unknown slug. GET /blog/:slug → { status, ok, data, error }
ActionTry it
Get capability tools
Every platform capability offered on this deployment and the tools each adds to an agent that enables it, as feature, tool name and tool description. GET /capability-tools → { status, ok, data, error }
ActionTry it
Get channel access requests
List organization early-access requests as { requests }. GET /channel-access-requests → { status, ok, data, error }
ActionTry it
Get conversations
One page of the organization’s conversations, most recent first, each with its agent, identity and latest state; with q, a full-text search of message content returning matching conversations with snippets. An unknown filter is refused, not ignored. GET /conversations → { status, ok, data, error }
ActionTry it
Get conversations by phone
One conversation thread by identity (phone number, email, or prefixed id like tg:123): its messages with their trace spans, the agent, blocked and takeover state, whether a turn is running, background tasks, and the peer for inter-agent threads. Long threads are large; page the messages with limit and offset. GET /conversations/:phone → { status, ok, data, error }
ActionTry it
Get conversations contact card
Read the contact card set specifically for this conversation, and whether dynamic contact cards are available. The conversation must belong to this organization; provide agentId or apiKey to disambiguate shared identities. GET /conversations/:phone/contact-card → { status, ok, data, error }
ActionTry it
Get conversations count
How many conversations match the same filters as the conversation list, and how many of those need a person: { all, needsReview }. GET /conversations/count → { status, ok, data, error }
ActionTry it
Get dashboard
Home-page summary for the signed-in user: their name, email, role and orgs (with the active one), headline stats (active conversations, messages today, average response time, total turns) and every agent in the org. GET /dashboard → { status, ok, data, error }
ActionTry it
Get deployment
Read one deployment in the current organization.
ActionTry it
Get evals cases
List the organization agent’s persisted eval cases, including generated, user-authored and regression cases. Returns cases without generating anything; a new agent may have an empty suite. GET /evals/cases → { status, ok, data, error }
ActionTry it
Get evals cases versions
Read the generated eval suite’s version history for an agent in the organization. Returns versions. GET /evals/cases/versions → { status, ok, data, error }
ActionTry it
Get evals diffs
Compare two organization eval runs as their results arrive. Returns diff with both arms’ scores and case-level fail-to-pass or pass-to-fail changes; use the run ids returned when starting a version comparison. GET /evals/diffs → { status, ok, data, error }
ActionTry it
Get evals runs by id
Read one organization eval run’s status, candidates, rankings and per-case results with trace ids. Marks stale runs as failed before reading and annotates outdated when the agent behavior changed. GET /evals/runs/:id → { status, ok, data, error }
ActionTry it
Get evals runs cases
Read the full persisted case records referenced by an organization eval run, including archived cases. Returns cases for inspecting or restarting a historical run; the case records may have been edited since it ran. GET /evals/runs/:id/cases → { status, ok, data, error }
ActionTry it
Get evals runs latest
Read the agent’s latest eval run, optionally limited to single-model, sweep or real-channel runs. Returns run or null, with outdated when the current agent behavior differs from its recorded fingerprint. GET /evals/runs/latest → { status, ok, data, error }
ActionTry it
Get evals runs transcript
Read an organization real-channel eval run’s recorded message transcript. Returns messages with caseId, inbound/outbound direction, text and ISO timestamp; runs without channel messages return an empty list. GET /evals/runs/:id/transcript → { status, ok, data, error }
ActionTry it
Get evals suggestions
Find behaviors worth testing from the agent’s recorded production traffic in the chosen window, with trace counts, error rates, models and existing suite coverage. Returns a sampled summary with suggestions; no model is called and no paid plan is required. GET /evals/suggestions → { status, ok, data, error }
ActionTry it
Get identities
List the org’s active messaging identities — SMS lines, Telegram, WhatsApp, Slack, Teams, X and AgentMail — with the agent each is attached to and its routing status. Slack rows include each workspace install, the install URL when the app is distributable, and the home channel; Teams rows include tenant installs and the messaging endpoint. GET /identities → { status, ok, data, error }
ActionTry it
Get identities phone calls forwardable lines
List the agent's iMessage lines and whether each can forward its calls: `{ lines: [{ number, support, connected }] }`. `support` is `instant`, `ticket` (the provider sets it up within a few days), `shared` (not possible) or `unsupported` (request it instead). GET /identities/phone-calls/forwardable-lines → { status, ok, data, error }
ActionTry it
Get identities phone calls numbers
Search Telnyx for available US local voice numbers in an area code or state, or anywhere in the US when neither is given. Returns `{ numbers: [{ phoneNumber, locality, state, monthlyCost }] }`; nothing is reserved or bought. GET /identities/phone-calls/numbers → { status, ok, data, error }
ActionTry it
Get identities phone calls voices
List the ElevenLabs voices a phone line can use: `{ voices: [{ id, name, description, previewUrl, labels }] }`. GET /identities/phone-calls/voices → { status, ok, data, error }
ActionTry it
Get identities slack channels
Per workspace that installed this Slack identity, the channels its bot is already in — the channel ids an escalation route may name — plus the scopes the install is missing, or an error when its token cannot list channels. GET /identities/:id/slack/channels → { status, ok, data, error }
ActionTry it
Get identities slack manifest
Build the Slack app manifest for a new identity and a prefilled api.slack.com app-creation URL, plus the events, interactivity and OAuth redirect URLs to configure and the bot scopes to grant. GET /identities/slack/manifest → { status, ok, data, error }
ActionTry it
Get identities slack operators
List the Slack users allowed to press this identity’s consequential buttons — taking a conversation over, replying to a customer, running an offered action. Org owners and admins whose Slack email matches are always allowed and need no entry. GET /identities/:id/slack/operators → { status, ok, data, error }
ActionTry it
Get identities slack signing probe by probe id
Check a Slack signing-secret probe: pending until a signed event from Slack arrives, then proven or failed. Returns the probe status and its channel. GET /identities/slack/signing-probe/:probeId → { status, ok, data, error }
ActionTry it
Get identities slack teams
For a Slack app that routes customer workspaces to different agents, list each workspace: its team id, the bound agent (or null), whether the app is installed there, and when it was last messaged while unbound. GET /identities/:id/slack/teams → { status, ok, data, error }
ActionTry it
Get identities teams manifest
The Teams app manifest for a connected Teams identity — it names that identity’s Microsoft App ID. Returns `{ messagingEndpoint, manifest }`, or the upload-ready zip package with icons when `format=zip`. GET /identities/teams/manifest → { status, ok, data, error }
ActionTry it
Get identities teams setup
Return the messaging endpoint an Azure Bot registration must point at before a Teams identity can be connected. GET /identities/teams/setup → { status, ok, data, error }
ActionTry it
Get identities whatsapp kapso available
Read platform Kapso configuration as { available }. GET /identities/whatsapp/kapso/available → { status, ok, data, error }
ActionTry it
Get identities x oauth available
Whether “Log in with X” can be offered — an X OAuth client is configured — plus which app-level credentials the platform already has, the redirect URI to register on the X app, and the hosted eval driver’s handle when one exists. GET /identities/x/oauth/available → { status, ok, data, error }
ActionTry it
Get identities x oauth callback
The redirect X sends the user back to after authorizing. Exchanges the code, checks the required scopes and offline access, resolves the account, and parks the grant for the connect call; it redirects to the channels page rather than answering JSON. GET /identities/x/oauth/callback → { status, ok, data, error }
ActionTry it
Get identities x oauth status
Poll a started X login: returns the authorization’s status and, once complete, the X account it resolved to — the input the connect call needs as `oauthState`. GET /identities/x/oauth/status → { status, ok, data, error }
ActionTry it
Get insights
Org analytics over a date range: daily activity, totals (messages, conversations, turns, errors, tokens, average latency), turns and messages per agent, task outcomes and response times. GET /insights → { status, ok, data, error }
ActionTry it
Get invitations
List the organization’s invitations, newest first, with email, role, status (pending, accepted or revoked) and expiry. Owners and admins only. GET /invitations → { status, ok, data, error }
ActionTry it
Get keys
List the organization’s API keys: name, prefix (never the full secret), scopes, bound agent, and last-used, expiry, revocation and creation times. GET /keys → { status, ok, data, error }
ActionTry it
Get keys usage
Daily usage of one API key, aggregated by day, by route and method, and by agent, with accepted vs rejected (revoked/expired) counts and first/last seen times. Requests still arriving after revocation (`attemptsAfterRevocation`) are the signal it is still in use. No per-request detail. GET /keys/:id/usage → { status, ok, data, error }
ActionTry it
Get lines
List the organization’s phone lines with channel, line type, routing status and message, activation time, and the agent each is assigned to. GET /lines → { status, ok, data, error }
ActionTry it
Get lines contact card
Read the contact card iMessage shows for a line — first and last name, photo URL, and whether the provider has applied it — straight from the line’s provider, plus the line’s channel. GET /lines/:number/contact-card → { status, ok, data, error }
ActionTry it
Get lines price
The display price of a new iMessage line (`price`, e.g. "$5", and `priceDetail`, e.g. "per month"); both are null when billing cannot be reached. GET /lines/price → { status, ok, data, error }
ActionTry it
Get lines providers
Status of each iMessage line provider (Photon, Linq, Blooio) as shown on the status page: today’s `state`, reachability, last check, 90-day uptime and daily history. The response includes `defaultProvider`; when set, it identifies the forced purchase provider and `components` is empty. GET /lines/providers → { status, ok, data, error }
ActionTry it
Get lines provisioning by id
Progress of one of the organization’s line purchases: `status` (pending, provisioning, action_required, requested, completed or error), the `phoneNumber` once ready, the Blooio `actionUrl` when the purchase needs completing, or the `error`. GET /lines/provisioning/:id → { status, ok, data, error }
ActionTry it
Get live snapshot
What the org’s agents are doing right now: the agents, turns from the last five minutes plus any still running with their steps, outcomes and new people from the last half hour, and today’s numbers. GET /live/snapshot → { status, ok, data, error }
ActionTry it
Get loops
List the organization’s scheduled loops, newest first: agent, the person it runs for, schedule, enabled and auto-paused state, last status and error, last and next run, and run count. GET /loops → { status, ok, data, error }
ActionTry it
Get marketplace capabilities
Browse published (reviewed) marketplace capabilities, one listing per capability with its newest version and the other versions available to pin, plus whether the caller is a capability reviewer. GET /marketplace/capabilities → { status, ok, data, error }
ActionTry it
Get marketplace mine
List the capabilities this organization has published itself, in every review state (draft, pending review, published, rejected). GET /marketplace/mine → { status, ok, data, error }
ActionTry it
Get marketplace reviews
List the capabilities awaiting publication review. Allowlisted capability reviewers and platform admins only. GET /marketplace/reviews → { status, ok, data, error }
ActionTry it
Get mcp brand lookup
Resolve the brand behind an MCP server endpoint URL — its name, domain, description and logo. Returns `{ brand: null }` when nothing is known. GET /mcp/brand-lookup → { status, ok, data, error }
ActionTry it
Get mcp catalog
Search the approved integrations.sh catalog of connectable services (name, domain, description, kind, icon, categories). With `q` the registry’s ranked search answers; without it the whole index comes back, optionally filtered by kind. Use an entry’s domain with the catalog-domain lookup to get its endpoints. GET /mcp/catalog → { status, ok, data, error }
ActionTry it
Get mcp catalog by domain
Resolve one catalogued domain into its connectable surfaces — MCP endpoints, OpenAPI/GraphQL specs and CLIs — with each surface’s URL, transports and the credential alternatives it needs. GET /mcp/catalog/:domain → { status, ok, data, error }
ActionTry it
Get mcp clients
List this user’s organization OAuth connections as { connections }. GET /mcp-clients → { status, ok, data, error }
ActionTry it
Get members
List the members of the connected organization with their role and last sign-in, plus the caller’s own user id and role and whether the caller may manage members. GET /members → { status, ok, data, error }
ActionTry it
Get models
Search the models an agent can run on — ids, context window, price per million tokens, and tool, reasoning and vision support. Resolve a model the user names here before setting it on an agent. GET /models → { status, ok, data, error }
ActionTry it
Get notifications
The signed-in user’s daily digest preference (`digestOptOut`) and the timezone the digest is sent in (null when none was reported). GET /notifications → { status, ok, data, error }
ActionTry it
Get org mcp integrations
List the organization-owned MCP integrations: endpoint, auth mode, header names (never values), which groups and agents each is granted to, and every materialised agent’s connection status and last error. GET /org-mcp-integrations → { status, ok, data, error }
ActionTry it
Get orgs
List every organization the signed-in user belongs to, with their role and logo in each, plus the id of the currently active organization. GET /orgs → { status, ok, data, error }
ActionTry it
Get orgs brand logo
Look up the brand logo URL for a domain. Returns the normalized domain and `logoUrl` (null when none is found). GET /orgs/brand-logo → { status, ok, data, error }
ActionTry it
Get orgs current
Read the active organization’s id, name, logo domain and logo URL, and whether the caller may edit it (`canManage`). GET /orgs/current → { status, ok, data, error }
ActionTry it
Get orgs current build recommendations
Read the agent ideas Home shows for the active organization: up to three items (title, description, channel, prompt) the organization wrote itself, else generated from its saved website domain, else a generic set. `source` says which (custom, website or generic). GET /orgs/current/build-recommendations → { status, ok, data, error }
ActionTry it
Get outcomes
Recorded business outcomes — sales, bookings, qualified leads — newest first, with totals per kind for the same agent, person and time scope. GET /outcomes → { status, ok, data, error }
ActionTry it
Get payments
Read the organization’s Grosso payments account: connected, pending a signup code, or not connected; test or live; and whether Grosso still accepts its key. Never returns the key. GET /payments → { status, ok, data, error }
ActionTry it
Get payments payees
Read the organization’s newest 100 Grosso payees (businesses it collects payments for): owner email, business name, status, whether it can be charged, and the one agent it belongs to (null when unassigned). GET /payments/payees → { status, ok, data, error }
ActionTry it
Get people
Directory of the people behind the organization’s conversations, most recently seen first, with the total matching count. GET /people → { status, ok, data, error }
ActionTry it
Get people by person id
One person: their identities across channels, conversations, loops, connected accounts, and their confirmed outcome count and value. GET /people/:personId → { status, ok, data, error }
ActionTry it
Get provider accounts
List the organization’s own Photon accounts connected to Keiki: label, Photon project id, connection status (`active`, or `error` with a message when Keiki’s webhook or credentials stopped working) and the lines imported from each. GET /provider-accounts → { status, ok, data, error }
ActionTry it
Get sandbox providers
Read sandbox provider availability as { providers }. GET /sandbox-providers → { status, ok, data, error }
ActionTry it
Get sandbox scripts
List the org’s sandbox scripts with the agents running each and its newest run — where a script that fails every boot shows up. GET /sandbox-scripts → { status, ok, data, error }
ActionTry it
Get sandbox scripts by id
Read one sandbox script, including its body, targets, timeout and env secret names. GET /sandbox-scripts/:id → { status, ok, data, error }
ActionTry it
Get sandbox scripts runs
Every sandbox run of one script across its agents, newest first — status, exit code and output tails. Use it to see why a script fails. GET /sandbox-scripts/:id/runs → { status, ok, data, error }
ActionTry it
Get sandbox scripts versions
Version history of a sandbox script, newest first — each entry carries the commit sha to read it at. GET /sandbox-scripts/:id/versions → { status, ok, data, error }
ActionTry it
Get sandbox scripts versions by sha
Read a sandbox script’s body as of one version, by the sha from its version history. GET /sandbox-scripts/:id/versions/:sha → { status, ok, data, error }
ActionTry it
Get secrets
List the organization’s developer secret names with created/updated times and any host binding; values are never returned. Setting or deleting a secret is not available as a call — use the manage_secret card. GET /secrets → { status, ok, data, error }
ActionTry it
Get signals
Built-in quality signals across every agent in the org, or one agent: events, share of classified users, change against the previous period, and average mood. GET /signals → { status, ok, data, error }
ActionTry it
Get skills
List the org’s skills and knowledge documents (metadata, no body) with the agents each is attached to. GET /skills → { status, ok, data, error }
ActionTry it
Get skills by id
Read one skill or document, including its full body. GET /skills/:id → { status, ok, data, error }
ActionTry it
Get skills presets
List the built-in preset skills (id, capability, name, description, kind) and whether the org already added each. Install one with POST /skills, sourceType "preset". GET /skills/presets → { status, ok, data, error }
ActionTry it
Get skills versions
Version history of a skill, newest first — each entry carries the commit sha to read it at. GET /skills/:id/versions → { status, ok, data, error }
ActionTry it
Get skills versions by sha
Read a skill’s body and metadata as of one version, by the sha from its version history. GET /skills/:id/versions/:sha → { status, ok, data, error }
ActionTry it
Get sso
Read SSO connection, domains, eligibility and platform configuration. Owners/admins only. GET /sso → { status, ok, data, error }
ActionTry it
Get trace exporters
List the destinations the org ships finished turns to — format, endpoint, header names (never values), enabled, attached agent ids, last delivery status — plus the formats available. GET /trace-exporters → { status, ok, data, error }
ActionTry it
Get traces
List one page of the org’s agent turns (traces), newest first, with status, model, tokens and error. Any query key not listed here is a 400. The page has no total; use /traces/count for that. GET /traces → { status, ok, data, error }
ActionTry it
Get traces by trace id
One turn with all its spans (every model step and tool call with full input/output). A turn with big tool payloads is large: try /traces/:traceId/summary first, or page the spans with limit/offset. GET /traces/:traceId → { status, ok, data, error }
ActionTry it
Get traces count
Count the turns matching the same filters as /traces; returns { total }. Any other query key is a 400. GET /traces/count → { status, ok, data, error }
ActionTry it
Get traces summary
What one turn did, in a small response: the reply it sent, every tool call with clipped input/output, and the spans that errored. Start here when debugging a turn. GET /traces/:traceId/summary → { status, ok, data, error }
ActionTry it
Get wallet programme
Whether agent wallets are offered on this deployment: each payment provider with whether it is configured and in test mode, plus overall configured/available flags. Wallet tools and routes do not exist when none is. GET /wallet/programme → { status, ok, data, error }
ActionTry it
Get webhook deliveries
List 50 outgoing consent/test webhook deliveries in this org, newest first. Legacy handoff and channel webhooks are not included. GET /webhook-deliveries → { status, ok, data, error }
ActionTry it
Get webhook deliveries by id
Read one org-owned outgoing delivery, its immutable event payload and the latest 100 attempts. GET /webhook-deliveries/:id → { status, ok, data, error }
ActionTry it
Get webhooks
List the org’s webhook endpoints (id, name, url) — the targets an escalation route can post to. GET /webhooks → { status, ok, data, error }
ActionTry it
Get webhooks by id
One webhook endpoint: its name, url and the events it subscribes to. GET /webhooks/:id → { status, ok, data, error }
ActionTry it
List agent templates
List the agent templates available for deployments.
ActionTry it
List channel types
List the channel types and required deployment fields.
ActionTry it
Patch identities group chat
Patch a Telegram identity’s group-chat behaviour — only the fields sent change; omitted fields keep their stored value. Returns the merged group-chat settings. Telegram identities only. PATCH /identities/:id/group-chat → { status, ok, data, error }
ActionTry it
Patch identities phone calls voice
Set a phone line's voice: answering model, ElevenLabs voice and delivery, first message, language, turn-taking, persona and tone. Omitted fields fall back to defaults. Returns the saved voice. PATCH /identities/phone-calls/:id/voice → { status, ok, data, error }
ActionTry it
Patch members by id
Change one member’s role, or their display name — one or the other per call. Changing a role, or anyone else’s name, is owners and admins only; you cannot change your own role. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. PATCH /members/:id → { status, ok, data, error }
ActionTry it
Patch notifications
Update the signed-in user’s daily digest settings; only the fields sent change. Returns the resulting preferences. An unrecognised timezone is rejected. PATCH /notifications → { status, ok, data, error }
ActionTry it
Patch orgs current
Rename the active organization, change its logo domain, or upload a logo; only the fields sent change. Changing the domain re-fetches the logo from it unless a logo is uploaded in the same call. Owners and admins only. Returns the updated organization. PATCH /orgs/current → { status, ok, data, error }
ActionTry it
Patch people by person id
Set the fields an operator owns on a person, patching only those sent: displayName, primaryEmail, notes. An omitted or blank name or email leaves the current value; notes are replaced by what is sent. Returns the person. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. PATCH /people/:personId → { status, ok, data, error }
ActionTry it
Patch sso
Update IdP metadata/configuration and defaultRole, enforce, activate; returns SSO settings. Eligible owners/admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. PATCH /sso → { status, ok, data, error }
ActionTry it
Post agent groups
Create an agent group, optionally with its members, and return its id. Owners and admins only. Names are unique per organization (409 if taken). With no hub every member asks every other; with one hub, spokes see and ask only the hub, the hub asks its spokes and alone keeps the group note. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /agent-groups → { status, ok, data, error }
ActionTry it
Post agent groups end
Stop a whole group: end every live peer thread among its members and cancel their background tasks. Returns how many threads were ended and tasks cancelled. Owners and admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /agent-groups/:id/end → { status, ok, data, error }
ActionTry it
Post agentmail domains
Add a custom email domain for agent inboxes. Returns the domain with the DNS records to publish; it can host inboxes once AgentMail has verified them. A domain already added by this or another org is refused. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /agentmail/domains → { status, ok, data, error }
ActionTry it
Post agents
Create an agent and return its id. Either send a full config (name, model and systemPrompt are required; everything else defaults), or send `template` to provision a complete template agent (only name, model, reasoningEffort and lineNumber are read alongside it) — that also returns `missing_secrets` still to set. POST /agents → { status, ok, data, error }
ActionTry it
Post agents a2a
Turn the agent’s A2A endpoint and Agent Card on or off. Returns the updated A2A view. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /agents/:id/a2a → { status, ok, data, error }
ActionTry it
Post agents actions decide
Approve (run) or decline one pending agent action and return the updated action; when its inputs changed since it was proposed it comes back superseded with what changed. Owners and admins only. Actions needing speaker scopes must be decided from Slack or the conversation instead. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /agents/:id/actions/:actionId/decide → { status, ok, data, error }
ActionTry it
Post agents actions groups decide
Approve or decline every pending action sharing a group id (from GET /agents/:id/actions), in order; approving stops at the first one that fails or is superseded. Returns each action’s outcome. Owners and admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /agents/:id/actions/groups/:groupId/decide → { status, ok, data, error }
ActionTry it
Post agents api gateway
Turn the agent’s OpenAI/Anthropic-compatible API on or off, and/or switch passthrough mode (the caller’s own prompt leads each turn) — send at least one. Passthrough needs the gateway on. Returns the updated gateway view. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /agents/:id/api-gateway → { status, ok, data, error }
ActionTry it
Post agents by id
Update an agent from its GET /agents/:id/edit config with your changes applied. Patches: an omitted field, or an omitted features toggle, keeps its stored value, but skillIds, sandboxScriptIds, sandboxEnvSecrets and userSecrets each replace their whole list. features.operatorHandoff hands work to a person in Slack and needs escalationRoutes.slack naming a connected install. Pass expectedUpdatedAt to refuse the save if someone else changed the agent. SDK-managed (local) agents are refused. POST /agents/:id → { status, ok, data, error }
ActionTry it
Post agents capabilities
Install a marketplace capability version on an agent, or re-pin an installed one to another version. Owners and admins only; the version must be published or the org’s own. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /agents/:id/capabilities → { status, ok, data, error }
ActionTry it
Post agents capabilities enabled
Enable or disable an installed capability, keeping its pinned version. Owners and admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /agents/:id/capabilities/:publisherRef/:slug/enabled → { status, ok, data, error }
ActionTry it
Post agents evals channel hosted
Start a real-channel eval run in which a platform-owned identity messages the agent over a real channel; returns the run id (watch it at /watch/:runId), the target handle and the cases run or excluded. Needs a paid plan, a configured driver and an active line for that channel; one channel run per agent at a time. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /agents/:id/evals/channel/hosted → { status, ok, data, error }
ActionTry it
Post agents fine tune start
Record a fine-tune request for the agent’s current version; refused until enough graded conversations exist. Owners and admins only. Training is scheduled separately, and repeating it returns the request already on file. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /agents/:id/fine-tune/start → { status, ok, data, error }
ActionTry it
Post agents harness redeploy
Replace a Maritime agent’s Keiki VM with a fresh one built from the current runner and return the new VM status. Only for agents on the maritime harness. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /agents/:id/harness/redeploy → { status, ok, data, error }
ActionTry it
Post agents harness signing secret
Rotate an external harness credential; returns { signingSecret }. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /agents/:id/harness/signing-secret → { status, ok, data, error }
ActionTry it
Post agents lines
Give an agent another phone line. A line belongs to one agent: a line another agent holds is refused; move it instead. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /agents/:id/lines → { status, ok, data, error }
ActionTry it
Post agents lines move
Move a line this agent holds to another agent in one step. A refused destination leaves the line where it was. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /agents/:id/lines/:number/move → { status, ok, data, error }
ActionTry it
Post agents loops clear history
Reset a loop: zero its run count and last status, delete its run history, and drop the snapshot it diffs against, so the next run starts fresh. Returns whether a snapshot was cleared. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /agents/:id/loops/:loopId/clear-history → { status, ok, data, error }
ActionTry it
Post agents loops enabled
Pause or resume one of the agent’s scheduled loops. Resuming recomputes its schedule, so a one-shot loop whose time has passed cannot be resumed. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /agents/:id/loops/:loopId/enabled → { status, ok, data, error }
ActionTry it
Post agents loops run
Run one of the agent’s scheduled loops now, through the same pipeline as a scheduled run. Returns whether it was queued or skipped, plus any capabilities that blocked it. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /agents/:id/loops/:loopId/run → { status, ok, data, error }
ActionTry it
Post agents mcp presets connect
Connect (or reconnect) one MCP preset on the agent’s Executor daemon. Returns the preset; when it needs OAuth its status asks for authorization and it carries the URL to open. Needs the Sandbox and MCP capabilities. POST /agents/:id/mcp-presets/:presetId/connect → { status, ok, data, error }
ActionTry it
Post agents mcp presets share
Share an agent’s MCP connection with the organization: its credential moves to an org-owned integration granted to the given groups and agents, then installed on each. Owners and admins only; a per-user-auth or already-shared connection cannot be shared. Returns the integration. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /agents/:id/mcp-presets/:presetId/share → { status, ok, data, error }
ActionTry it
Post agents mcp presets status
Re-check an MCP preset’s connection on the daemon without restarting OAuth — poll it after the user authorizes. Returns the preset with its current status. POST /agents/:id/mcp-presets/:presetId/status → { status, ok, data, error }
ActionTry it
Post agents notice
Tell the agent something in its own conversations and let it decide whether the people in them hear about it — how you notify everyone using an agent (an outage, a capability going away, a migration). The text is never delivered: the agent reads it as a platform notice and decides, in each contact’s history and tone, whether and how to say anything, so write it for the agent and never draft the message itself. With `phone` it runs one conversation and returns its outcome (ran, queued for its next turn, or unavailable); without it it fans out over every conversation (newest first, bounded) and answers 202 with how many it started — report that as started, not delivered. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /agents/:id/notice → { status, ok, data, error }
ActionTry it
Post agents quality apply
Apply a drafted fix to the agent’s live configuration (a new agent version). Only a draft that passed its evaluation gate can be applied. A model-change draft needs the model chosen in the body; a prompt draft takes no body. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /agents/:id/quality/:proposalId/apply → { status, ok, data, error }
ActionTry it
Post agents quality dismiss
Dismiss a drafted fix, together with the failure cluster it came from so the same symptom is not suggested again. POST /agents/:id/quality/:proposalId/dismiss → { status, ok, data, error }
ActionTry it
Post agents quality settings
Change the agent’s turn-review settings. Patches only the fields sent; a threshold outside its bounds is refused. POST /agents/:id/quality/settings → { status, ok, data, error }
ActionTry it
Post agents rollouts percent
Set the share of conversations an open rollout serves its staged config to, and resume it when paused. Returns the updated rollout; promote it to reach 100. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /agents/:id/rollouts/:rolloutId/percent → { status, ok, data, error }
ActionTry it
Post agents rollouts promote
Make the rollout’s staged config live for every conversation, recorded as a new version, and end the rollout. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /agents/:id/rollouts/:rolloutId/promote → { status, ok, data, error }
ActionTry it
Post agents rollouts rollback
End the rollout and send every conversation back to the live config. The live config is not changed. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /agents/:id/rollouts/:rolloutId/rollback → { status, ok, data, error }
ActionTry it
Post agents template import
Re-apply the agent’s own template, at the template’s current version, onto its configuration (e.g. when GET /agents/:id reports the template as outdated). Fails when the agent was not made from a template. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /agents/:id/template/import → { status, ok, data, error }
ActionTry it
Post agents tools
Save a new authored code tool as a disabled draft (saving under the name of an existing authored tool rewrites that draft). Test it, then enable it to publish. Returns the saved tool. POST /agents/:id/tools → { status, ok, data, error }
ActionTry it
Post agents tools enable
Publish an authored tool to the agent so it can call it, or unpublish it with enabled false. Returns the tool. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /agents/:id/tools/:toolId/enable → { status, ok, data, error }
ActionTry it
Post agents versions restore
Roll the agent back to the configuration saved at a version sha, recorded as a new version whose sha is returned. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /agents/:id/versions/:sha/restore → { status, ok, data, error }
ActionTry it
Post billing credits purchase
Buy usage credits for the organization; owners and admins only. Charges the saved payment method, or returns a checkoutUrl to share with the user when none is saved; a null URL means the credits were added. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /billing/credits/purchase → { status, ok, data, error }
ActionTry it
Post billing plans attach
Attach the selected plan to the organization; owners and admins only. Read available plans first and choose one with selfServe: true; other plans are operator-assigned and attach rejects them. Share any returned checkoutUrl with the user; a null URL means no checkout is needed. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /billing/plans/attach → { status, ok, data, error }
ActionTry it
Post billing release line
Release an active phone line owned by the organization and unassign it from agents; owners and admins only. For the last dedicated Photon line, also stop its subscription; returns ok, billingStopped and numberReturned to distinguish billing cancellation from number return. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /billing/release-line → { status, ok, data, error }
ActionTry it
Post channel access requests
Submit { channel, purpose, expectedVolume }; returns { request }. Configurers only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /channel-access-requests → { status, ok, data, error }
ActionTry it
Post conversations block
Block one identity for this conversation’s agent, so its messages no longer trigger agent replies. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /conversations/:phone/block → { status, ok, data, error }
ActionTry it
Post conversations end
End an inter-agent (agent:) thread: settle its pending asks and refuse new ones until the identity is unblocked. Returns how many asks were settled. 409 for a customer conversation. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /conversations/:phone/end → { status, ok, data, error }
ActionTry it
Post conversations messages
Send a message the contact reads as the agent’s, over the conversation’s own channel, and renew the takeover. Only while the conversation is taken over (409 otherwise, or on a channel with no outbound send). Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /conversations/:phone/messages → { status, ok, data, error }
ActionTry it
Post conversations steer
Run a turn on this conversation as if the contact had written the text, and return what the agent would answer with its trace id. Nothing reaches the contact: delivery tools are refused and the answer is kept in the dashboard, marked as never sent. Works during a takeover as a draft; refused (409) while the agent is mid-turn. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /conversations/:phone/steer → { status, ok, data, error }
ActionTry it
Post conversations takeover
Hold one conversation for a human for 30 minutes: incoming messages are stored for them to answer and no agent turn runs. Returns the takeover with its expiry; sending a message renews it. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /conversations/:phone/takeover → { status, ok, data, error }
ActionTry it
Post evals cases
Add a user-authored eval case to an agent in the organization and return the saved case. Name and prompt are required, and a duplicate active case name is rejected; code-tool check inputs are free-form JSON. POST /evals/cases → { status, ok, data, error }
ActionTry it
Post evals cases generate
Re-author and persist the agent’s generated eval suite from its current configuration, keeping user and regression cases and recording a suite version. Requires a paid plan and, for passthrough agents, a captured caller spec; returns cases and spends OpenRouter credits using the resolved platform or BYOK key. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /evals/cases/generate → { status, ok, data, error }
ActionTry it
Post evals diffs
Start paid-plan evals comparing two different readable config versions of the organization’s agent against up to 12 existing cases. Requires a nonempty suite and no in-flight comparison; returns base_run_id and candidate_run_id to poll the comparison, and spends credits running both versions. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /evals/diffs → { status, ok, data, error }
ActionTry it
Post evals runs
Start a paid-plan eval run on the agent’s saved suite, or selected case ids including archived cases; generation occurs only if the saved suite is empty. Omit candidates to use the configured model, or supply up to 20 distinct models for comparison; returns run_id to poll, or a conflict with the already running run_id. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /evals/runs → { status, ok, data, error }
ActionTry it
Post evals runs cancel
Request cooperative cancellation of an organization eval run that is still running; the runner checks between cases. Returns ok true only when a running run was marked cancelled, otherwise false; no body is read. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /evals/runs/:id/cancel → { status, ok, data, error }
ActionTry it
Post evals suggestions create
Create regression eval cases from selected production suggestions by replaying recorded conversation evidence in a mocked world, without model calls. Read suggestions for the same window first; returns caseIds, skipped suggestion ids and the saved cases, skipping stale or already covered suggestions. POST /evals/suggestions/create → { status, ok, data, error }
ActionTry it
Post identities agentmail
Create an AgentMail email identity attached to an agent — a new inbox on the managed domain or one of the org’s verified custom domains, or an existing inbox the org already owns — and register its inbound webhook. An agent can have only one AgentMail inbox. POST /identities/agentmail → { status, ok, data, error }
ActionTry it
Post identities agentmail reconnect
Rotate an AgentMail identity’s inbox-scoped API key — mints a replacement, stores it, deletes the old key, and when attached re-registers the inbound webhook. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /identities/:id/agentmail/reconnect → { status, ok, data, error }
ActionTry it
Post identities agentmail verify
Verify an AgentMail identity: re-fetches the inbox, and when the identity is attached to an agent (re)registers the inbound webhook. Updates and returns the routing status. POST /identities/:id/agentmail/verify → { status, ok, data, error }
ActionTry it
Post identities attach
Route an identity to an agent: registers the channel’s inbound webhook or subscription and marks the identity routing-active. Requires a paid plan. Returns the resulting routing status. POST /identities/:id/attach → { status, ok, data, error }
ActionTry it
Post identities detach
Stop routing an identity to its agent: tears down the channel’s webhook or subscription and leaves the identity unattached, so inbound messages stop being answered. The identity itself is kept. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /identities/:id/detach → { status, ok, data, error }
ActionTry it
Post identities phone calls
Buy a voice number in the area code or state (or any US number when neither is given) and connect it to the agent for inbound phone calls (charges the org). Returns the identity id and the purchased number. With `forwardFrom`, the agent's iMessage line forwards its calls to a hidden voice number instead and stays the public number; `forwarding` is `active`, or `pending` while the line's provider sets it up. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /identities/phone-calls → { status, ok, data, error }
ActionTry it
Post identities phone calls forwarding requests
Ask the team to set up call forwarding by hand for an iMessage line whose provider cannot do it automatically. POST /identities/phone-calls/forwarding-requests → { status, ok, data, error }
ActionTry it
Post identities phone calls voices preview
Speak a short text (at most 300 characters; defaults to the first message) in a voice. Returns MP3 audio. POST /identities/phone-calls/voices/preview → { status, ok, data, error }
ActionTry it
Post identities slack
Create a Slack identity from a manually-built Slack app: bot token and signing secret are required, the OAuth client pair makes the app installable by other workspaces. Optionally attached to an agent. Fails if the app is already connected. POST /identities/slack → { status, ok, data, error }
ActionTry it
Post identities slack install link
Mint a signed Slack install link that binds the installing workspace to an agent. Owners and admins only. An unpinned link can be reused by any workspace that is not bound yet. POST /identities/:id/slack/install-link → { status, ok, data, error }
ActionTry it
Post identities slack install links revoke
Stop every install link made so far for a customer-workspace Slack app, without touching its signing secret. Installed workspaces keep working. Owners and admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /identities/:id/slack/install-links/revoke → { status, ok, data, error }
ActionTry it
Post identities slack operators
Add one Slack user to a Slack identity’s operator allowlist without touching the other entries. Owners and admins only. Returns the resulting list. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /identities/:id/slack/operators → { status, ok, data, error }
ActionTry it
Post identities slack reconnect
Replace a Slack identity’s credentials with a fresh bot token for the same Slack app — the token must resolve to the app already connected — plus its signing secret. The pasted token replaces that workspace’s install; a blank OAuth client keeps the one already stored. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /identities/:id/slack/reconnect → { status, ok, data, error }
ActionTry it
Post identities slack signing probe
Store a candidate Slack signing secret and have the bot post a setup instruction into a channel; a human then mentions the bot there so Slack’s signed event proves the secret. Returns the probe id and whether the instruction was posted. POST /identities/slack/signing-probe → { status, ok, data, error }
ActionTry it
Post identities slack validate
Check pasted Slack app credentials before storing them: the workspace and bot the token belongs to, the scopes the install is missing, the channels the bot can see, whether the OAuth client pair is real, and whether the app is already connected. The signing secret cannot be checked here — only a signed request from Slack proves it. POST /identities/slack/validate → { status, ok, data, error }
ActionTry it
Post identities slack verify
Re-check every workspace install of a Slack identity against Slack: refreshes workspace names and granted scopes, and retires an install whose token Slack has revoked. Returns per-install results; a revoked everywhere token raises a routing fault on the identity. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /identities/:id/slack/verify → { status, ok, data, error }
ActionTry it
Post identities teams
Create a Microsoft Teams identity from an Azure Bot registration — the App ID and a client secret are validated by minting a Bot Framework token. Optionally attached to an agent. Returns the identity id, routing status, and the messaging endpoint the registration must point at. POST /identities/teams → { status, ok, data, error }
ActionTry it
Post identities telegram
Create a Telegram bot identity, optionally attached to an agent; the inbound webhook is registered only when an agent is attached. Returns the new identity id, its routing status, and the resolved bot. Fails if the bot is already connected anywhere. POST /identities/telegram → { status, ok, data, error }
ActionTry it
Post identities telegram validate
Resolve a Telegram bot token through getMe and return the bot’s id, username and first name, without persisting anything. Use it to preview which bot a token belongs to before creating the identity. POST /identities/telegram/validate → { status, ok, data, error }
ActionTry it
Post identities whatsapp
Create a WhatsApp Cloud API identity, optionally attached to an agent; webhook subscription runs only when an agent is attached. Returns the identity id, routing status, the resolved number, and the webhook URL to point the Meta app at. Fails if the number is already connected. POST /identities/whatsapp → { status, ok, data, error }
ActionTry it
Post identities whatsapp kapso
Connect { phoneNumberId, agentId? } from this organization’s Kapso signup; returns { ok, id, routingStatus, number }. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /identities/whatsapp/kapso → { status, ok, data, error }
ActionTry it
Post identities whatsapp kapso own
Connect { apiKey, phoneNumberId, agentId? } from a tenant Kapso account; returns { ok, id, routingStatus, number }. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /identities/whatsapp/kapso/own → { status, ok, data, error }
ActionTry it
Post identities whatsapp kapso start
Mint an organization Kapso signup link; returns { url } for the user to open. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /identities/whatsapp/kapso/start → { status, ok, data, error }
ActionTry it
Post identities whatsapp validate
Resolve a Meta access token + phone number id to the WhatsApp number they belong to — display number and verified name — plus the webhook URL to register, without persisting anything. POST /identities/whatsapp/validate → { status, ok, data, error }
ActionTry it
Post identities x
Create an X account identity for public mentions and direct messages, from a completed “Log in with X” grant. The app credentials sign and register webhooks — when the platform app authorized the login they may be omitted. Attaching an agent requires a paid plan. Fails if the account is already connected. POST /identities/x → { status, ok, data, error }
ActionTry it
Post identities x oauth start
Start an X OAuth 2.0 login that mints a refreshable user token for an X identity. Returns the state and the authorize URL to send the user through; the callback parks the grant for the connect call. POST /identities/x/oauth/start → { status, ok, data, error }
ActionTry it
Post identities x reconnect
Replace the authorization of an X identity whose token X revoked, using a completed “Log in with X” for the same account. The grant is consumed once; app credential fields update only when sent. POST /identities/:id/x/reconnect → { status, ok, data, error }
ActionTry it
Post identities x validate app credentials
Check that an X app API key + API key secret pair can mint the app-only bearer token that registers the X webhook, without storing anything. Returns the bearer inspection, or `{ ok: false, error }`. POST /identities/x/validate-app-credentials → { status, ok, data, error }
ActionTry it
Post invitations
Invite someone to the organization by email. Owners and admins only. Emails the invitation when email is configured; otherwise returns the accept link (`acceptUrl`) to share. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /invitations → { status, ok, data, error }
ActionTry it
Post invitations resend
Resend a pending invitation with a fresh accept link and a renewed expiry. Owners and admins only. Returns whether the email was sent, and the accept link (`acceptUrl`) to share when email is not configured. POST /invitations/:id/resend → { status, ok, data, error }
ActionTry it
Post keys
Create an API key, optionally bound to a single agent of this organization. Returns the full key once — it cannot be read again. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /keys → { status, ok, data, error }
ActionTry it
Post keys revoke
Revoke an API key so every later request with it is rejected. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /keys/:id/revoke → { status, ok, data, error }
ActionTry it
Post keys rotate
Rotate an API key: the old secret stops working and the new full key is returned once. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /keys/:id/rotate → { status, ok, data, error }
ActionTry it
Post lines contact card
Set the first and last name iMessage shows for a line, applied through its provider (Photon or Linq) and returned as the saved card. Owners and admins only. The photo is unchanged; uploading one is only possible from the dashboard. POST /lines/:number/contact-card → { status, ok, data, error }
ActionTry it
Post lines preview
Quote a new iMessage line with a coupon applied, without charging, redeeming the coupon or provisioning: returns subtotal, total and currency (major units) and the next invoice, which may include other plans. Owners and admins only; an invalid coupon is a 400. POST /lines/preview → { status, ok, data, error }
ActionTry it
Post lines purchase
Buy a new iMessage line for the organization from Photon (default) or Blooio, optionally with a coupon and connected straight to an agent. Owners and admins only. Returns a `checkoutUrl` the user must open when payment details are needed. Otherwise the selected provider provisions the line; when `DEFAULT_IMESSAGE_PROVIDER` is set, it overrides the request body’s `provider`. A Photon line is provisioned immediately and its `phoneNumber` returned; a Blooio purchase may instead return `status: "provisioning"` (or `"action_required"` with an `actionUrl`) and a `requestId` to follow with GET /lines/provisioning/:id. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /lines/purchase → { status, ok, data, error }
ActionTry it
Post lines request
Ask the Keiki team for a Linq line. Owners and admins only. Nothing is charged; the request is recorded, the team is alerted and fills it by hand later. Returns the `requestId`; repeating while one is open returns the same request. Returns 409 without recording when `DEFAULT_IMESSAGE_PROVIDER` forces Photon or Blooio. POST /lines/request → { status, ok, data, error }
ActionTry it
Post lines verify
Check one line’s inbound webhook registration with its provider and its stored routing status; it changes nothing except promoting a degraded or unserved status back to active when the webhook checks out. Returns the expected webhook URL, whether a webhook is registered and matches the stored id, and findings such as missing webhooks, id mismatches, channel/provider mismatches or no outbound line selection. It does not prove outbound delivery works, and the stored channel is configuration intent (possibly the `linq` default), not proof. POST /lines/:number/verify → { status, ok, data, error }
ActionTry it
Post marketplace mine submit
Submit one of the organization’s own capability versions for marketplace publication review; it is emailed to the review inbox. Owners and admins only. A bundle zip and notes can only be attached from the dashboard upload form, not through this call. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /marketplace/mine/:capabilityId/submit → { status, ok, data, error }
ActionTry it
Post marketplace reviews by capability id
Approve (publish) or reject a capability that is awaiting review, with an optional note to the publisher. Allowlisted capability reviewers and platform admins only; fails unless the capability is pending review. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /marketplace/reviews/:capabilityId → { status, ok, data, error }
ActionTry it
Post members transfer ownership
Make another member the owner of the organization; the current owner becomes an admin. Owner only. Find the member’s user id with the member list first. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /members/transfer-ownership → { status, ok, data, error }
ActionTry it
Post org mcp integrations sync
Re-run the reconciliation of an org MCP integration into every granted agent now, in the background. Owners and admins only. Returns the integration as it stands; poll the list for each agent’s new status. POST /org-mcp-integrations/:id/sync → { status, ok, data, error }
ActionTry it
Post payments identity verification
Start payment identity verification; returns { url }. Owners/admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /payments/identity-verification → { status, ok, data, error }
ActionTry it
Post payments key
Connect Grosso using { apiKey }; returns { account } without its key. Owners/admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /payments/key → { status, ok, data, error }
ActionTry it
Post payments live
Switch payments to live mode; returns { account }. Owners/admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /payments/live → { status, ok, data, error }
ActionTry it
Post payments payees
Create { ownerEmail, agentId, maxAmount, dailyAmount } (integer cents); returns { payee, onboardingUrl }. Owners/admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /payments/payees → { status, ok, data, error }
ActionTry it
Post payments payees by id
Assign a payee with { agentId }; returns { payee }. Owners/admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /payments/payees/:id → { status, ok, data, error }
ActionTry it
Post payments signup
Start Grosso signup with { email, name }; returns { account }. Owners/admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /payments/signup → { status, ok, data, error }
ActionTry it
Post payments verify
Verify signup with { code, acceptTerms: true }; returns { account }. Owners/admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /payments/verify → { status, ok, data, error }
ActionTry it
Post sandbox scripts
Create a sandbox script: a "build" script runs once per sandbox (again when its body changes), a "startup" script on every boot. Creating does not attach it to any agent. Returns the created script. POST /sandbox-scripts → { status, ok, data, error }
ActionTry it
Post skills
Create a skill or knowledge document — always a new one. Owners, admins and members only. sourceType picks the source: "file" (filename + inline content, .md/.markdown/.txt, up to 2 MB), "url" (a raw .md/.txt URL), "website" (a web page imported as a document), "npx" (an `npx skills add` GitHub source) or "preset" (presetId from /skills/presets; 409 if already added). Returns the created skill. Does not attach it to any agent. POST /skills → { status, ok, data, error }
ActionTry it
Post sso domains
Register { domain }; returns { domain } with its verification record. Eligible owners/admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /sso/domains → { status, ok, data, error }
ActionTry it
Post sso setup link
Create a single-use SSO setup credential; returns { setupUrl, expiresAt }. Eligible owners/admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /sso/setup-link → { status, ok, data, error }
ActionTry it
Post sso test
Start an SSO test; returns { url } for the user to open. Eligible owners/admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /sso/test → { status, ok, data, error }
ActionTry it
Post subagent tasks cancel
Cancel one running or suspended background task an agent delegated; the agent is told it was cancelled. Returns cancelled: false when the task had already finished. Owners and admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /subagent-tasks/:id/cancel → { status, ok, data, error }
ActionTry it
Post trace exporters
Add a destination that receives every finished turn of its attached agents. Owners, admins and members only. The endpoint must resolve to a public address at delivery. Returns the created exporter. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /trace-exporters → { status, ok, data, error }
ActionTry it
Post traces abort
Ask a running turn to stop; it halts at its next step. 409 if the turn is not running. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /traces/:traceId/abort → { status, ok, data, error }
ActionTry it
Post traces retry
Re-run a failed turn with its original input, or re-run its linked loop; returns which was queued. 409 with a reason when it cannot (not failed, conversation moved on or busy, taken over by a person, input gone). Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /traces/:traceId/retry → { status, ok, data, error }
ActionTry it
Post webhook deliveries resend
Owners/admins only: queue a resend using the original event ID/payload. The destination must be enabled and no attempt can be in progress. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /webhook-deliveries/:id/resend → { status, ok, data, error }
ActionTry it
Post webhooks
Create a named webhook endpoint. 409 if the name is taken. Returns the webhook. POST /webhooks → { status, ok, data, error }
ActionTry it
Post webhooks signing secret
Rotate a destination webhook credential; returns { signingSecret }. Owners/admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /webhooks/:id/signing-secret → { status, ok, data, error }
ActionTry it
Post webhooks test
Queue a synthetic webhook delivery; returns { eventId } with status 202. Owners/admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. POST /webhooks/:id/test → { status, ok, data, error }
ActionTry it
Put agent groups by id
Update an agent group: patches only the name and description sent, and when members is sent replaces the whole member list with it (read the group first to keep existing members). Owners and admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. PUT /agent-groups/:id → { status, ok, data, error }
ActionTry it
Put agent groups note
Replace the whole group note with a new body and return the new revision; restoring a past version is a PUT of that version’s body. Read the note first to keep what still holds. Owners and admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. PUT /agent-groups/:id/note → { status, ok, data, error }
ActionTry it
Put agents harness connection
Configure an external harness with { mode: "inbox"|"push", dispatchUrl? }; returns { ok }. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. PUT /agents/:id/harness/connection → { status, ok, data, error }
ActionTry it
Put agents live
Pause an agent or put it back live. A paused agent keeps its channels and settings and replies to no one until it is live again. Returns { live }. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. PUT /agents/:id/live → { status, ok, data, error }
ActionTry it
Put agents mcp presets
Replace the agent’s whole MCP preset list (at most 10): presets left out are disconnected and removed, and editing a preset’s name, endpoint, kind, auth mode or headers reconnects it — read the agent’s current presets first and send them back with their ids. Needs the Sandbox and MCP capabilities. Returns the reconciled presets. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. PUT /agents/:id/mcp-presets → { status, ok, data, error }
ActionTry it
Put agents mcp presets tool permissions
Replace the allow/ask/deny choices for one MCP’s tools and push them to the daemon. Both fields are required and replace what is stored, so read the tools first and carry over the stored choices. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. PUT /agents/:id/mcp-presets/:presetId/tool-permissions → { status, ok, data, error }
ActionTry it
Put agents note
Replace the whole working note body (read it first to keep what matters) and return the new revision. Owners and admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. PUT /agents/:id/note → { status, ok, data, error }
ActionTry it
Put agents public profile
Publish or unpublish the agent’s public profile page (which shows its name and phone line) and set its tagline. Owners and admins only. Replaces both: an omitted tagline is cleared, so read the profile first to keep it. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. PUT /agents/:id/public-profile → { status, ok, data, error }
ActionTry it
Put agents tools by tool id
Update an authored tool. Patches only the fields sent; any save disables the tool again until it is re-enabled. Returns the saved tool. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. PUT /agents/:id/tools/:toolId → { status, ok, data, error }
ActionTry it
Put agents user connections
Replace the agent’s personal provider definitions with the list sent (read GET first and keep the providers you are not changing) and return the saved list. A non-empty list needs the MCP and sandbox features on; SDK-managed agents are refused. This only offers providers: each user still authorizes through connect_account, and no shared credential is copied. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. PUT /agents/:id/user-connections → { status, ok, data, error }
ActionTry it
Put agents wallet
Update the agent’s wallet, creating it if needed: spending limits, approval threshold, merchant allowlist, user-card policy, payment method, payment provider (only while nothing is connected) or pause/resume. Owners and admins only. Patches only the fields sent. Returns the saved wallet. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. PUT /agents/:id/wallet → { status, ok, data, error }
ActionTry it
Put billing auto top up
Turn auto top-up on or off; owners and admins only. While enabled, credits are bought with the saved payment method whenever the balance falls below threshold. threshold and quantity are required even when disabling. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. PUT /billing/auto-top-up → { status, ok, data, error }
ActionTry it
Put conversations contact card
Set the name, organization, title and optional photo for this conversation’s contact card. The conversation must belong to this organization; provide agentId or apiKey to disambiguate shared identities. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. PUT /conversations/:phone/contact-card → { status, ok, data, error }
ActionTry it
Put evals cases by id
Replace an eval case’s editable fields and adopt it as user-authored so regeneration preserves it; read the current case first, since omitted history, world, channel, stubs, features and ordinary checks reset to defaults. Omitted seed and checks.effects/executes are preserved; returns the saved case, with free-form JSON code-tool inputs. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. PUT /evals/cases/:id → { status, ok, data, error }
ActionTry it
Put identities slack events webhook
Send a customer-workspace Slack app’s slack.installed, slack.uninstalled and slack.unbound_message events to one of the org’s saved webhooks (GET /webhooks lists them). Owners and admins only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. PUT /identities/:id/slack/events-webhook → { status, ok, data, error }
ActionTry it
Put identities slack operators
Replace a Slack identity’s operator allowlist wholesale. Owners and admins only. This overwrites the whole list — read it first with GET /identities/:id/slack/operators, and prefer POST / DELETE to change one person without risking a concurrent edit. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. PUT /identities/:id/slack/operators → { status, ok, data, error }
ActionTry it
Put identities slack teams by team id
Route one customer Slack workspace to an agent, replacing any previous binding. Owners and admins only. Returns the workspace list. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. PUT /identities/:id/slack/teams/:teamId → { status, ok, data, error }
ActionTry it
Put org mcp integrations credentials
Replace the authentication headers of a header-authenticated org MCP integration; every granted agent is reconnected with the new values in the background. Owners and admins only. Returns the updated integration. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. PUT /org-mcp-integrations/:id/credentials → { status, ok, data, error }
ActionTry it
Put org mcp integrations grants
Replace the full set of groups and agents an org MCP integration is granted to — anything not listed loses access, so read the current grants first. The integration’s source agent always stays granted. Owners and admins only. Granted agents are reconciled in the background; returns the updated integration. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. PUT /org-mcp-integrations/:id/grants → { status, ok, data, error }
ActionTry it
Put orgs current build recommendations
Replace the Home ideas the active organization wrote itself; they take precedence over the website-based ones. Send all of them (at most three) every time; an empty list clears them. Owners and admins only. Returns the ideas Home now shows. PUT /orgs/current/build-recommendations → { status, ok, data, error }
ActionTry it
Put sandbox scripts by id
Edit a sandbox script in place, patching only the fields sent; id, attachments and version history survive. Changing a build script’s body makes already provisioned sandboxes run it again. Returns the updated script. PUT /sandbox-scripts/:id → { status, ok, data, error }
ActionTry it
Put secrets by name
Store an encrypted secret with { value }; returns { ok, name }, never its value. Configurers only. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. PUT /secrets/:name → { status, ok, data, error }
ActionTry it
Put skills by id
Edit a skill in place, patching only the fields sent; its id, slug, agent attachments and version history survive. A content that is a full SKILL.md sets the name and description from its frontmatter, and an explicit name or description wins over it. Returns the updated skill. PUT /skills/:id → { status, ok, data, error }
ActionTry it
Put trace exporters by id
Update a destination — replaces name, format, endpoint, enabled and agentIds, so read it first and send them all; headers are kept when omitted. Owners, admins and members only. Returns the updated exporter. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. PUT /trace-exporters/:id → { status, ok, data, error }
ActionTry it
Put webhooks by id
Update a webhook endpoint — replaces both name and url, so send both. Returns the webhook. Destructive — deleting, spending money, or changing who has access. Show the user what it will do and get their confirmation first. PUT /webhooks/:id → { status, ok, data, error }
ActionTry it
Switch organization
List the organizations this account belongs to, or switch this connection to one of them. Every later call then acts in that organization, with the role the user has there. Call it with no organization to see the choices first.
ActionTry it
Talk to agent
Send one message to one of this organization's agents as a real user, and read what it replies. The turn runs on the agent's own configuration and tools; nothing is delivered to a real line, but every other tool it calls (MCP integrations, sandbox, browser, loops, payments) runs for real and can change or delete data outside this conversation, and fresh: true erases this client’s earlier thread with the agent. Use it to find out what an agent actually answers, to reproduce something a user reported, or to check an agent after changing it. Call it again to continue the conversation. A turn that uses tools takes a minute or more and reports progress while it runs; if it is still going when your client would give up, it answers with a jobId instead of a reply — that is not a failure, read the rest with check_job. Files the agent sends come back in `files` with download URLs valid for up to 15 minutes, never past their retention expiry; `check_job` on a finished job re-issues them while they are retained.
ActionTry it
How the Keiki MCP integration works
The Keiki MCP integration connects your Dench AI CRM directly to Keiki MCP, so agents can read and act on your Keiki MCP data as part of everyday work — answering questions in chat, keeping your CRM in sync, and running automations without anyone copying data between tools.
336 actions are available for agents to invoke on your behalf. Every call runs through Keiki MCP's own authorization, scoped to the account you connect.
Set up Keiki MCP in Dench
- 1
Sign in to your Dench workspace and open Integrations.
- 2
Find Keiki MCP and click Connect — you'll authorize access through Keiki MCP's own sign-in flow. No API keys or code required.
- 3
Ask an agent to use Keiki MCP in chat, or call it from an automation.
- 4
Manage or disconnect the connection any time from workspace settings.
Frequently asked questions
How does the Keiki MCP integration work with Dench?
The Dench Keiki MCP integration connects your AI CRM to Keiki MCP, so AI agents can work with your Keiki MCP data as part of chats, automations, and CRM workflows. You connect your account once, and every agent in your workspace can use it — governed by your workspace permissions.
What actions can AI agents perform with Keiki MCP via Dench?
The Keiki MCP integration currently exposes 336 actions, including Cancel job, Check job, Current context, Delete agent groups by id, Delete agents by id, and Delete agents capabilities by slug. Agents invoke them on your behalf from chat or from automations.
Do I need to write code to connect Keiki MCP to Dench?
No. You connect Keiki MCP from your Dench workspace using Keiki MCP's own sign-in and authorization flow — no API keys to copy, no glue code to maintain.
Is the Keiki MCP integration secure?
Connections are authorized through Keiki MCP's own authentication flow, and Dench stores only the authorization needed to act on your behalf. You can review and disconnect the Keiki MCP connection from your workspace settings at any time.