Niches · Code & developer platforms
Fix a bug from a recorded session or bug-bash run
No company's docs show a full fit yet. The listings below do the job with a limit, or not at all.
What counts as done
The reproduction steps, console error and failing request are extracted into the agent's context, and a fix is proposed against the file the stack trace names.
Measured, for our own tests: extraction takes one call per artefact. No docs we read state these figures, so they don't count against a listing here.
How agents ask for it: read this Jam and fix the bug
, bug-bash the preview deploy for this PR
, what did the console say when the user hit the error
Does the job, with a limit 2
It does the job, but something in the docs or a user's account will get in an agent's way. The limit is named in each row.
| Listing | Why |
|---|---|
| Jam MCP C Jam | getConsoleLogs with logLevel and limit, getNetworkRequests with bodies, getUserEvents, getVideoTranscript and getFrames each return one artefact per call, and a personal access token works as a Bearer header for CI; but there is no read-only mode and deleteJam and deleteFolder are exposed with no rate limits or error docs. source |
| Momentic MCP server D Momentic | start_session takes a reachable app or GitHub PR, get_session returns the report and finding details, list_findings pages bugs; but the server takes OAuth only and rejects API keys, has no read-only mode, and the page does not say whether console errors or failing requests are returned. source |
Doesn't do this job 19
Listings in the category that do something else, with the reason.
| Listing | Why |
|---|---|
| amCharts 5 MCP Server B amCharts | No recording or debug tools. |
| Azure DevOps MCP Server B Microsoft Corporation | No recording or browser tools. |
| Context7 BB Upstash | No recording or debug tools. |
| Entire BB Entire Inc | No recording or browser tools. |
| Git (MCP reference server) C Model Context Protocol, a Series of LF Projects, LLC | No recording or debug tools. |
| GitHub MCP Server BB GitHub, Inc. | No recording or browser tools. |
| GitLab MCP Server B GitLab Inc. | No recording or browser tools. |
| GitMCP E Ido Salomon (GitMCP open-source project) | No recording or debug tools. |
| graft C Vishisht Choudhary (individual) | No recording or debug tools. |
| Graphify BB Graphify Labs | No recording or debug tools. |
| Microsoft Learn MCP Server D Microsoft Corporation | No recording or debug tools. |
| ModusToolbox AI Assistant D Infineon Technologies AG | No recording or debug tools. |
| REA BB morluto | Browser observation and HAR capture tools inspect a live or saved target, not a Jam-style bug recording with repro steps. source |
| Ref D Ref Software, Inc | No recording or debug tools. |
| Salesforce DX MCP Server C Salesforce, Inc. | No recording or debug tools. |
| Sourcegraph MCP Server B Sourcegraph, Inc. | No recording or debug tools. |
| Superpowers B Prime Radiant (no legal form or registration found on its pages); the MIT LICENSE names Jesse Vincent | systematic-debugging is a method, not a recording reader. |
| TokenEconomizer B Microsoft Corporation (repository in the microsoft GitHub organisation with Microsoft's SECURITY.md); co-authors at UC Berkeley | No recording or debug tools. |
| Void B Cloudflare, Inc. (VoidZero) | No recording tools; logs come from the deployed worker only. |
What stops agents finishing
- no read-only mode and delete tools exposed (Jam)
- tool list, limits and errors not published (Momentic)
Around it
Before: observability.triage-error. After: code.ticket-to-pr.
Where this comes from
Weight 3 of 100 in Code & developer platforms: No first-hand account found for Jam MCP or Momentic; weight is a placeholder resting on vendor pages. Treat as unverified.
The same as data: /api/v1/use-cases/code.debug-from-recording.json, or the MCP tool get_use_case with id code.debug-from-recording. Other jobs in this category: /api/v1/niches?category=code.
For companies
Do agents find, use and choose your tools?
An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.