What it does
errorwatch gives an agent or a shell access to a core subset of the Sentry API. It can confirm which user or token is authenticated, list organizations, projects and releases, and search the issue stream using Sentry's own query syntax. It can also open an issue with its latest event, change an issue's status, assignment and flags, and read or add comments.
It suits workflows where an agent needs to triage Sentry errors, for example checking the newest event on an issue or marking one resolved or assigned. The same bin runs as a CLI, and its mcp subcommand starts a stdio MCP server that MCP clients can configure. The container image defaults to the MCP server, and passing CLI arguments switches it to CLI mode.
Running it requires a Sentry auth token in the SENTRY_AUTH_TOKEN environment variable, sent as a Bearer token and never printed in output. If the variable is missing, every call fails with an error naming it. CLI exit codes are 0 on success, 1 when an API call fails, and 2 for usage errors. Failed API calls are reported as tool errors with the HTTP status and Sentry's detail message when one is provided.
Limitations and requirements
Requires a Sentry auth token in SENTRY_AUTH_TOKEN; every call fails without it. Covers only a core subset of the Sentry API. Documentation search, Seer AI autofix, profiling, traces and spans search, DSN and project creation, other admin operations, and organization statistics beyond issues are out of scope.
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/errorwatch
Organizations, projects, releases, and issues through the Sentry API.
v0.1.0As shown on registry.modelcontextprotocol.io when we last read it