Skip to content

Audit: README/AGENTS.md code-example accuracy #33

Description

@smiggleworth

Purpose

Audit task, not a bug report — produces a verification table, doesn't fix anything.

Procedure

  1. Extract every fenced code block from README.md and AGENTS.md.
  2. Verify each against current behavior, including the static-asset serving section (Static asset routing silently shadows dynamic routes with dotted paths; duplicated content-type parsing #28 already shows a gap between documented and actual behavior there).
  3. File a gap issue for anything stale.

Definition of Done

  • Table: doc file | section | ✅ verified accurate / 🔗 gap filed.
  • Every runnable-looking code block has a row.
  • No row left unmarked.

Audit results (2026-08-24)

Document Runnable fence Verification Result
README.md #1 (sh) Install command matches package name.
README.md #2 (ts) createNodeHandler usage matches public declarations and handler tests.
README.md #3 (ts) listen options match public declarations and bind tests.
README.md #4 (ts) WebSocket route/listen example matches server and Node WebSocket contracts.
README.md #5 (ts) serve example now documents assets.exclude and is covered by #35.
README.md #6 (ts) Public bind/allowed host example matches host-validation tests.
README.md #7 (ts) MCP stdio example matches connectMcpStdio declarations/tests.
AGENTS.md none No fenced runnable example exists.

The README behavior gap identified by #28 is corrected in #35; no other stale example was found.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions