Runheed MCP server

• Registry namecom.runheed/runheedverified in the MCP registry

The registry name sits under runheed.com. The registry only lets a publisher use that namespace after proving control of the domain through DNS or HTTP.

• Stateanswerschecked 2026-11-02
• Tools9
• Server calls itselfrunheed
• Server version0.1.0
• Protocol version2025-06-18

On 2026-11-02 the server answered initialize and tools/list without a sign-in. First answer in 260 ms.

• Whether its tools do the product's job has not been judged yet.
• Checked on 1 of the 7 days it takes to judge reliability.

Said by rules from the measurements above. The likely first call is a model's pick from the tool list; read its description before relying on it.

• INITE MCP scoreafter a weeka score needs 7 daily checks: 1 so far
• Answeredmeasuring1 of the 7 daily checks it takes to print a share
• Median first answer260 ms
• Space in an agent's context≈ 1,030 tokensthe definitions of all its tools (9), which an agent carries on every turn
• Tool definitionsA · 100 of 100
• Typed results0 of 9 toolsdeclared outputSchema: results come typed, read once

• Every tool passes all six checks.

A program checks the server once a day: initialize and tools/list, no model. Tokens are the definitions' length at about four characters a token. The grade averages six checks of every tool's name, description, schema and annotations; it says how well the tools are described, not whether they work. The score weighs what was measured and nothing else: answering (25%), doing the product's job (25%), definitions (20%), context (10%), use (10%) and how readily a client connects (10%). A part that was not measured is left out, and a hidden instruction caps the score at 40. The state of vendors' MCP servers: method and numbers →

• runheed_search_errors

  Searches the workspace's error groups by text (type or message), status and project. Returns each matching group with its status, type, message, event count, last-seen time and ID.

• runheed_get_error

  Returns one error group in full: message, status, counts, first and last seen, environment, agent context, the top stack frames and the latest breadcrumbs.

• runheed_get_recent

  Returns the error groups seen most recently, newest first, with status, type, message, event count, last-seen time and ID.

• runheed_get_trends

  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.

• runheed_get_error_context

  Returns everything recorded around an error: every breadcrumb, platform, runtime, OS, environment, release, the affected user, the agent session and extra data.

• runheed_resolve_error

  Marks an error group resolved, with an optional note such as the fixing commit. The group reopens on its own if the error happens again. Returns a confirmation.

• runheed_ignore_error

  Marks an error group ignored, for known or expected errors, so new occurrences stop alerting. Returns a confirmation.

• runheed_unignore_error

  Moves an ignored error group back to unresolved, so new occurrences alert again. Returns a confirmation.

• runheed_get_project_health

  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.