Skip to content

chore(release): 0.7.1 - #121

Merged
theDakshJaitly merged 1 commit into
mainfrom
release/0.7.1
Aug 5, 2026
Merged

chore(release): 0.7.1#121
theDakshJaitly merged 1 commit into
mainfrom
release/0.7.1

Conversation

@theDakshJaitly

Copy link
Copy Markdown
Collaborator

Patch release carrying the agent guidance fix merged in #119. Documentation and tool-config changes only — no source, schema, or CLI behaviour changes.

Contents

  • package.json / package-lock.json -> 0.7.1
  • CHANGELOG.md entry

src/version.ts reads the version from package.json at runtime, so there is no second place to bump.

Why a release rather than waiting

The guidance in #119 only reaches users through the published package — templates/ ships via the files whitelist, and mex setup copies from it. Until it is published, the fix is only on GitHub.

Known limitation, stated in the changelog

Upgrading does not rewrite an existing .mex/ scaffold. Tool configs are copied at setup time (src/setup/index.ts), and there is no refresh command, so existing users keep the old ## Code Graph block. The changelog tells them to replace that section from templates/AGENTS.md.

Extending src/drift/checkers/tool-config-sync.ts to flag a stale Code Graph block against the shipped template would close this automatically and is worth a follow-up — that checker already compares tool configs to each other.

Verification

  • npm run build clean, including DTS and .wasm/schema asset copy
  • npx vitest run 369/369 passing (four consecutive full runs)
  • npm pack --dry-run: mex-agent-0.7.1.tgz, 38 files, 1.1 MB packed / 8.3 MB unpacked
  • Confirmed templates/AGENTS.md, templates/.tool-configs/*, dist/cli.js, and dist/schema.sql are all in the tarball, and that the shipped templates carry the new guidance text

One intermittent full-suite failure was observed twice earlier in the session and could not be reproduced across four subsequent runs; CI on #119 passed on both Node 22 and 24. test/cli.test.ts shells out to npm run build from inside the suite, which is a plausible race against parallel tests reading dist/. Worth a separate look; not treated as a blocker here.

Not included

#120 (README results framing, CODEOWNERS, eval fixture) is deliberately held back for a later release.

Patch release carrying the agent guidance fix from #119. Documentation and
tool-config changes only; no source or schema changes.

Note that upgrading does not rewrite an existing .mex/ scaffold — tool configs
are copied at setup time — so existing users need to replace the Code Graph
section manually. Called out in the changelog.
Copilot AI lite review requested due to automatic review settings August 5, 2026 08:11

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

Bumps the mex-agent package version to 0.7.1 and documents the patch release in the changelog, reflecting the previously merged agent-guidance updates from #119.

Changes:

  • Bump npm package version to 0.7.1 (package.json, package-lock.json).
  • Add a 0.7.1 entry to CHANGELOG.md describing the shipped tool-config guidance update and upgrade note.

Reviewed changes

Copilot reviewed 2 out of 3 changed files in this pull request and generated 1 comment.

File Description
package.json Updates published package version to 0.7.1.
package-lock.json Aligns lockfile/root package metadata to 0.7.1.
CHANGELOG.md Adds 0.7.1 release notes and guidance for existing scaffolds.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Comment thread CHANGELOG.md
### Changed
- Agent guidance in the shipped tool configs now describes when to use each graph command, rather than preferring graph commands over text search in all cases. `mex graph query` and `mex graph get` lead when a symbol name is known — they are exact and typically 200-500 output tokens. `mex graph scope` is positioned as a starting point for unfamiliar tasks, with an explicit note that it matches on words rather than meaning, so a task phrased in vocabulary the code does not use will return weak results.
- Agents are now told to fall back to Grep/Glob when a scope manifest does not contain what they need, and to rephrase a scope task at most once. The previous wording discouraged text search, which could lead an agent to spend additional calls expanding a manifest that was not going to answer the question.
- Applied to `AGENTS.md`, `CLAUDE.md`, `.cursorrules`, `.windsurfrules`, and `.github/copilot-instructions.md`.
@theDakshJaitly
theDakshJaitly merged commit d009454 into main Aug 5, 2026
3 checks passed
@theDakshJaitly
theDakshJaitly deleted the release/0.7.1 branch August 5, 2026 08:17
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.

2 participants