Skip to content

docs(#494): update logpoint support + attach adapterConfig docs to post-#469/#466 reality - #506

Merged
debugmcpdev merged 2 commits into
mainfrom
fix/494-stale-docs
Aug 27, 2026
Merged

docs(#494): update logpoint support + attach adapterConfig docs to post-#469/#466 reality#506
debugmcpdev merged 2 commits into
mainfrom
fix/494-stale-docs

Conversation

@debugmcpdev

Copy link
Copy Markdown
Collaborator

Summary

Fixes #494 — three user-facing descriptions still documented pre-#469 / pre-#466 behavior, two of them in the MCP tool schema itself (the text an agent reads to decide what to call).

Changes

  1. set_breakpoint logMessage schema description (src/server.ts): Ruby joins the exclusion list ("not by Java, .NET, or Ruby" — Ruby logpoints silently become pausing breakpoints; list_breakpoints then reports verified:true with the logMessage intact #469 made ruby fail fast), and mock joins the supported list, aligning the schema string with the tool-reference table.
  2. docs/tool-reference.md logpoint table: Ruby moves into the fails-fast row (with the Ruby logpoints silently become pausing breakpoints; list_breakpoints then reports verified:true with the logMessage intact #469 rationale); the "unknown — accepted with a warning; validated at launch" row now describes the general dynamically-loaded-adapter mechanism it actually is, instead of being pinned to Ruby.
  3. docs/tool-reference.md attach_to_process: removed the sentence claiming "adapterConfig is not applied to js-debug attach" (it contradicts adapterConfig: unknown keys are accepted silently — a typo of a supported key (pathMapping vs pathMappings) gets no warning #466localRoot/remoteRoot/sourceMaps/skipFiles/continueOnAttach do reach js-debug), and documented both halves of the warning contract: dropped keys named as ignored vs unrecognized keys forwarded as-is with an edit-distance did-you-mean.

All three were verified empirically in the issue against main @ 9a7327a. No behavior changes; grep confirms no test pins the old strings.

🤖 Generated with Claude Code

#469/#466 reality

- set_breakpoint's logMessage schema description now lists Ruby among the
  fail-fast adapters (#469) and mock among the supported ones, matching
  the tool-reference table.
- tool-reference logpoint table: Ruby moves to the fails-fast row; the
  "unknown -> validated at launch" row now describes the general
  dynamically-loaded-adapter mechanism it actually is.
- tool-reference attach_to_process: the "adapterConfig is not applied to
  js-debug attach" sentence contradicted #466 (js-debug attach honors
  localRoot/remoteRoot/sourceMaps/skipFiles/continueOnAttach); the
  warning contract now documents both outcomes — dropped keys named as
  ignored, unrecognized keys forwarded as-is with a did-you-mean.

Closes #494

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
@codecov

codecov Bot commented Aug 27, 2026

Copy link
Copy Markdown

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

@debugmcpdev
debugmcpdev merged commit 7465906 into main Aug 27, 2026
10 checks passed
@debugmcpdev
debugmcpdev deleted the fix/494-stale-docs branch August 27, 2026 01:03
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Docs: logpoint support table + logMessage schema still describe pre-#469 Ruby behavior; tool-reference contradicts #466 on js-debug attach adapterConfig

2 participants