What it does
cueport gives an MCP host or a terminal access to CueAPI, a coordination layer for agent work. It can create recurring (cron) or one-time cues, fire them on demand, list and inspect their executions, claim worker executions, keep claims alive with heartbeats, and report outcomes backed by evidence. A single binary provides both a command-line interface and a stdio MCP server, which exposes tools for each of these operations.
It suits cases where agent work is scheduled from an MCP host or a shell, or where a worker needs to claim tasks, extend its claim and record results. CLI commands print the upstream JSON payload and exit with 0 on success, 1 on a failed call, or 2 on a usage error. Pausing and resuming are PATCH updates that set the cue status. The claim-next command fetches the server-filtered claimable list, picks the oldest entry and claims it by id.
Set CUEAPI_API_KEY to a key from your CueAPI account; it is sent as a bearer token. Set CUEAPI_BASE_URL to target a self-hosted or other CueAPI-compatible host; otherwise calls go to https://api.cueapi.ai. The container image runs the MCP server by default, and passing CLI arguments runs it as a CLI instead.
Limitations and requirements
A CueAPI API key is required in CUEAPI_API_KEY; without it, calls fail before any request is sent. Upstream API errors also cause a failed call with exit code 1.
Safety
Our team has not reviewed this resource. Only automated rules checked it; resources with high-risk signals are not published. These are signals, not a guarantee that installing it is safe. Try it in a sandbox project first.
- Runs code or changes an assistant's behaviour
Sources
- Official MCP Registry first seen 13 Nov 2026 · last seen 13 Nov 2026 · its page on the source
io.github.CloudCorpRecords/cueport
CueAPI agent-work scheduling: cues, executions, claiming, heartbeats, and outcomes.
v0.1.0As shown on registry.modelcontextprotocol.io when we last read it