Can I use this with Claude Desktop?

Yes, add the cueapi entry to claude_desktop_config.json using the npx command.

Does it support claiming work as a worker?

Yes, the execution-lifecycle tools let agents list, claim, heartbeat, and report outcomes.

mcp

Verified

Schedule agent work and report outcomes via CueAPI.

MCP ServerAI & KnowledgeLocal (stdio) 2

View on GitHub

Updated 2026-06-16

What is the mcp MCP server?

cueapi-mcp exposes the full CueAPI surface inside any MCP host so agents can create, pause, fire, and update cues without leaving the conversation. It also supports the complete execution lifecycle including claiming work, heartbeats, and writing once outcomes with external evidence.

The server is distributed as @cueapi/mcp on npm and connects to either the hosted CueAPI service or a self-hosted instance using the provided API key.

Install & connect

Add this to your MCP client config. Pick your client below and copy.

{
  "mcpServers": {
    "mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@cueapi/mcp"
      ],
      "env": {
        "CUEAPI_API_KEY": "<YOUR_CUEAPI_API_KEY>",
        "CUEAPI_BASE_URL": "<CUEAPI_BASE_URL>"
      }
    }
  }
}

Package: @cueapi/mcp (npm)

Tools it exposes

15 tools your AI client can call once connected.

cueapi_create_cue— Create a recurring (cron) or one-time (`at`) cue

cueapi_list_cues— List cues, filter by status

cueapi_get_cue— Fetch details for a single cue

cueapi_fire_cue— Fire an existing cue immediately, optional payload override

cueapi_update_cue— Update a cue's mutable fields (name, schedule, callback, payload, etc.)

cueapi_pause_cue— Pause a cue so it stops firing

cueapi_resume_cue— Resume a paused cue

cueapi_delete_cue— Delete a cue permanently

cueapi_list_executions— List historical executions, filter by cue/status

cueapi_get_execution— Fetch a single execution by ID, with state + outcome

cueapi_list_claimable_executions— List unclaimed worker executions, filter server-side by task/agent

cueapi_claim_execution— Atomically claim a specific execution for processing

cueapi_claim_next_execution— Claim the next available execution (optional task filter)

cueapi_execution_heartbeat— Extend the claim lease on an in-flight execution

cueapi_report_outcome— Report write-once outcome with evidence (external ID / URL)

Example prompts

Once connected, try asking your AI client:

“Schedule a daily 9am job that posts a digest to my webhook.

“Show me the last five times cue_abc123 ran.

“Claim the next available execution for the digest task.

“Report that execution exec_123 completed with result URL https://example.com/report

Configuration

Environment variables this server needs.

Variable	Required	Description
`CUEAPI_API_KEY`secret	Yes	CueAPI API key. Generate at https://cueapi.ai
`CUEAPI_BASE_URL`	No	Override the CueAPI base URL for self-hosted instances. Defaults to https://api.cueapi.ai

Security & permissions

Requires CUEAPI_API_KEY (and optionally CUEAPI_BASE_URL) passed via environment variables to the stdio process; the key grants access to create, modify, and report on cues and executions in your CueAPI account.

What you can do with mcp

Daily digest scheduling

Create a cron cue that posts a summary to a webhook every morning at 9am.

Agent work handoff

Schedule follow-up tasks and later claim executions to continue processing across sessions.

Outcome reporting

Report execution results with external IDs or URLs so downstream systems receive proof of completion.

How to use mcp

1Install with npm install -g @cueapi/mcp or run via npx.
2Generate an API key at cueapi.ai.
3Add the server entry to your MCP host config with the npx command and CUEAPI_API_KEY env var.
4Optionally set CUEAPI_BASE_URL for self-hosted instances.
5Restart the host and start using the cueapi_* tools.

mcp: pros & cons

Pros

+Full coverage of cue and execution lifecycle tools in one package.
+Works with any stdio-capable MCP host including Claude Desktop, Cursor, and Zed.
+Supports both hosted and self-hosted CueAPI deployments.
+Write-once outcome reporting with external evidence links.

Cons

–Parameter schemas are not documented in the README so agents must discover inputs at runtime.
–All state lives in the remote CueAPI service; no local persistence.
–Requires a paid or self-hosted CueAPI account for production use.

Did you find this helpful?

Frequently asked questions

CUEAPI_API_KEY is mandatory; CUEAPI_BASE_URL is optional for self-hosting.

User reviews

Verified reviews from the community shape this listing's rating.

Loading reviews…

Promote mcp

Add this badge to your website, or share the tool.