Runheed
Server Details
Production errors your coding agent can read: search, stack traces, releases, resolve, setup.
- Status
- Healthy
- Uptime
- 100.0% over the last 24 hours
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-11-25
- URL
- Repository
- getrunheed/runheed-plugin
- GitHub Stars
- 0
- ChatGPT
- –
- Claude
- Runheed
TDQS
Scored across 13 tools
Most tools have clearly distinct purposes (resolve/ignore/unignore/merge are well separated, and project vs error resources are obvious). The only real overlap is between runheed_get_error, runheed_get_error_context and runheed_get_recent, since get_error already returns stack frames and breadcrumbs that get_error_context re-exposes; descriptions do enough to steer selection.
Every tool follows the same runheed_verb_noun snake_case pattern (create_project, get_error, search_errors, merge_errors, send_test_event). Verbs are used consistently for their action class (get/list/search for reads, resolve/ignore/unignore for status changes).
13 tools is well within the healthy range and each earns its place: project setup, error reading, triage actions, and aggregate health/trends views. There is no obvious filler or redundant endpoint.
The surface covers the full error-triage lifecycle (discover, inspect, resolve/ignore/unignore, merge, verify, plus health/trends and a test event) so agents won't hit dead ends in core workflows. Minor gaps remain: no project update/delete or key rotation, and no way to delete or permanently discard an error group.
Available Tools
13 toolsrunheed_create_projectCreate a projectAInspect
Creates a project in the workspace and returns its DSN: the value an app sets as the RUNHEED_DSN environment variable so that runheed.init() reports its errors to this project.
Inputs: name (required), slug, platform.
runheed_get_errorGet an errorARead-onlyInspect
Returns one error group in full: message, status, counts, first and last seen, environment, agent context, the top stack frames and the latest breadcrumbs. If the error came back after it was marked resolved, it also shows the note it was resolved with, so you can see what the last fix tried. For a resolved error it reports whether the fix held: events since, releases deployed since and whether the fixing commit is live.
Inputs: errorId (required).
runheed_get_error_contextGet error contextARead-onlyInspect
Returns everything recorded around an error: every breadcrumb, platform, runtime, OS, environment, release, the affected user, the agent session and extra data.
Inputs: errorId (required).
runheed_get_project_healthGet project healthARead-onlyInspect
Returns a 0 to 100 health score for a project (or the whole workspace), errors in the last 24 hours, the unresolved count, the error rate, recent issues and recommendations.
Inputs: projectId.
runheed_get_recentGet recent errorsBRead-onlyInspect
Returns the error groups seen most recently, newest first, with status, type, message, route, top stack frame, event count, last-seen time and ID.
Inputs: limit, projectId.
runheed_get_trendsGet error trendsARead-onlyInspect
Returns error statistics for the last 24 hours, 7 days or 30 days: total, new and resolved counts, the most frequent errors and the hourly error rate.
Inputs: period, projectId.
runheed_ignore_errorIgnore an errorAIdempotentInspect
Marks one error group ignored (errorId), or up to 100 at once (errorIds), for known or expected errors such as browser noise, so new occurrences stop alerting. Returns what changed.
Inputs: reason, errorId, errorIds.
runheed_list_projectsList projectsARead-onlyInspect
Lists the workspace's projects with each project's ID, slug, platform and DSN (the value an app sets as RUNHEED_DSN to report errors).
Inputs: no named parameters.
runheed_merge_errorsMerge duplicate errorsAInspect
Merges error groups that are the same bug into one: their events move to the target group (targetId) and new events with their fingerprints join it too. With groupByMessagePrefix, new events of the target's type whose message starts with that text also join it, for messages that differ only by an ID or a number. Returns what was merged.
Inputs: errorIds (required), targetId (required), groupByMessagePrefix.
runheed_resolve_errorResolve an errorAIdempotentInspect
Marks one error group resolved (errorId), or up to 100 at once (errorIds) for duplicates fixed by the same change, with an optional note such as the fixing commit. A group reopens on its own if the error happens again. Returns what changed.
Inputs: comment, errorId, errorIds.
runheed_search_errorsSearch errorsARead-onlyInspect
Searches the workspace's error groups by text (type, message, route, file or extra data), status, project and release (including only the errors a release introduced). Returns each matching group with its status, type, message, route, top stack frame, event count, last-seen time and ID.
Inputs: limit, query, status, release, +2 more.
runheed_send_test_eventSend a test eventAInspect
Sends one test error (RunheedTestEvent) to a project through the normal ingest path, which shows whether the project and its send key work. The test error appears in the recent errors and can be resolved like any other.
Inputs: projectId (required).
runheed_unignore_errorStop ignoring an errorAIdempotentInspect
Moves one ignored error group (errorId), or up to 100 (errorIds), back to unresolved, so new occurrences alert again. Returns what changed.
Inputs: errorId, errorIds.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
Production errors and incidents for your agent: stack traces, breadcrumbs, likely causes, triage.
Production errors of your Node.js and TypeScript apps, ready for your agent to fix.
Error tracking for small teams and their coding agents.
Production bugs with replay, network, console and stack, ranked for your coding agent to fix.
Related MCP Servers
- AlicenseAqualityBmaintenanceEnables turning pasted or logged crash output from Python, Node, Java, Go, Rust, and .NET into compact root-cause reports, mapping frames onto project code, fingerprinting and comparing traces, explaining exceptions, and redacting secrets over MCP.7MIT

Errata MCP Serverofficial
AlicenseAqualityFmaintenanceEnables AI coding agents to check PyPI and npm packages for sandbox-verified outdated-code corrections, search errors by text or traceback, and submit confirmation or dispute reports.397 npmMIT- AlicenseBqualityBmaintenanceEnables AI agents to search service logs, group errors by signature, compare a candidate run against a baseline for regressions, and correlate which problems from other services appear before or after an error. All results are computed by the tools rather than the language model, so agents can only explain findings the code produced.5MIT
- AlicenseAqualityCmaintenanceEnables coding agents to locally investigate and manage error tracking issues, including Laravel error summaries, paginated stack traces, triage, comments, and deletion.85 npm1MIT
Glama MCP Gateway
Add one secure layer between your agents and this server.