Repository navigation
Stage 1: tool annotations, homepage, log scrubbing, license, CI - #4
Merged
Merged
Conversation
- Tool annotations on all five tools (title + readOnlyHint/destructiveHint, idempotentHint, openWorldHint), required by the Anthropic and OpenAI directories. delete_transcript is destructive; get/summarize/understand are read-only; submit_transcript is open-world (AssemblyAI fetches the URL). - A short page at / for people who open the server in a browser: endpoint, Claude Code setup command (built from the request host), and the tools. AI apps keep using /mcp; nothing there changes. - logError scrubs URLs and long opaque tokens from error messages and stacks, so upstream error text can't carry customer values into logs. - MIT LICENSE, matching the other AssemblyAI SDK repos. - GitHub Actions CI on every PR and push to main: npm ci, typecheck, offline smoke tests, next build. - Smoke tests 43 (annotations) and 44 (log scrubbing): 101/101 passing. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
The scheme check used startsWith("localhost"), so a host like
localhost.example.com was shown as http://. Match the hostname exactly
(localhost or 127.0.0.1, any port).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Stage 1 of the MCP roadmap: what the Anthropic and OpenAI directories check for, plus basic repo hygiene. No change to how
/mcpbehaves for clients.Tool annotations
Every tool now has a
titleand annotations, which both directories require (the Anthropic portal flags tools withoutreadOnlyHint/destructiveHint):submit_transcriptget_transcriptsummarize_transcriptunderstand_transcriptdelete_transcriptHomepage at
/A short page for people who open the server in a browser (it was a 404): what it is, the endpoint, a Claude Code setup command built from the request host, and the five tools. AI apps keep using
/mcp.Log scrubbing
logErrornow strips URLs and long opaque tokens from error messages and stacks before logging. Upstream API error text is logged verbatim and can echo request values, so this keeps audio links, webhook targets and key-like strings out of Vercel logs. Transcript IDs are logged in their own field and are unaffected.License + CI
LICENSE, matching the AssemblyAI Node/Python SDKs and CLI.main:npm ci, typecheck, offline smoke tests,next build. Once this merges,maincan require this check (branch protection).Test plan
npm run typecheckcleannpm test: 101/101, including new test 43 (annotations: titles, delete destructive, reads read-only) and test 44 (log scrubbing)npm run buildproduces/,/mcp,/.well-known/oauth-protected-resource/renders the page,/mcpstill returns 401 without a key/shows the page,/mcpunchanged, live tests 22/22🤖 Generated with Claude Code