[wukongnotnull]

This PR MUST target the main branch. Active work is developed on
focused branches and merged directly into main after review.

Tracking: #53

Who is submitting this PR? (required)

Field	Value
Your model + version	Cursor Grok 4.6
Harness + version	Cursor cloud agent
All plugins installed	none beyond the Cursor cloud-agent environment (MCP namespaces present: cursor-cloud, cursor-subscriptions; unused for the bump itself)
Human partner who reviewed this diff	悟空非空也g — asked to 发版 after cleanup, with origin/main already containing merged #50 and #51

What problem are you trying to solve?

Manifests on main still say 6.4.0 while merged work after that tag is already on the default branch. Users who install by version, and bug reports that cite 6.4.0, cannot distinguish the post-tag commits. RELEASE-NOTES.md already had an Unreleased section for #37–#45; #50 (collaboration rules) and #51 (docs/prd.md) landed later and were not in a dated release heading. 悟空非空也g asked to 发版 after remote-branch cleanup (#52).

This is the same problem #39 solved for 6.4.0: ship the version number and notes together so harness install paths and the changelog match main.

What does this PR change?

Runs scripts/bump-version.sh 6.5.0 on the seven declared files (package.json, Claude/Cursor/Codex/Kimi plugin manifests, Claude marketplace plugin version, product-design.lock.json wukong_code_version). Renames ## Unreleased to ## v6.5.0 (2026-10-13), keeps the existing Unreleased bullets, and adds a Documentation section for #50 and #51 only. Updates the README Product Design version line and the docs/prd.md current-version line so the snapshot matches the tag.

Is this change appropriate for the core library?

Yes. This is core plugin release metadata and notes, not a new skill, not a third-party dependency, and not a rewrite of behavior-shaping skill text. The lock-file wukong_code_version field is outside the hashed Product Design import boundary (same as #39).

What alternatives did you consider?

• Patch 6.4.1. Rejected: Unreleased already records a user-visible feature (Chinese Codex routing, feat(hooks): route Chinese prompts in the Codex language router #41) plus CI/packaging/docs. Same reasoning as Release v6.4.0 #39 rejecting 6.3.1.
• Major 7.0.0. Rejected: no harness install incompatibility or removals.
• Bump without adding docs: add GitHub issue and Projects collaboration rules #50/docs: add Simplified Chinese product requirements document #51 to the notes. Rejected: those commits are on main and would otherwise vanish from the dated section.
• Invent extra features in the notes. Rejected by the release request; notes only describe landed PRs.
• Also create a GitHub Release / npm publish / Codex portal zip in this PR. Out of scope for the version-bump commit. After merge, the documented follow-up is an annotated v6.5.0 tag (as for 6.4.0). GitHub Releases exist for 6.2.x and 6.3.0 but not 6.4.0; npm has no wukong-code package. Those steps will be attempted after merge and reported on Ops: release v6.5.0 from current main #53 if credentials block them.

Does this PR contain multiple unrelated changes?

No. Every file is the version bump, the release-notes heading for that bump, or the two version strings the previous release PR also updated (README) plus the PRD version line that did not exist at 6.4.0.

Existing PRs

• I have reviewed all open AND closed PRs for duplicates or prior art
• Related PRs: Release v6.4.0 #39 (Release v6.4.0 — same path; this is the next cut, not a redo), docs: add Unreleased release-notes entry for #37–#45 #46 (Unreleased notes for ci: re-pin evals-static to a reachable harness commit #37–ci: gate pull requests on scripts/lint-shell.sh --all #45 — consumed here, not duplicated), Release v6.2.0 #4 / Release v6.2.1 #10 (older release PRs), docs: add GitHub issue and Projects collaboration rules #50 / docs: add Simplified Chinese product requirements document #51 (content being dated, not re-landed). No open duplicate release PR.

Difference from #39: that PR dated 6.4.0. This PR dates everything merged after that tag. It does not change skills or tests; #39 also had to stop hardcoding the PD test version, which is already done.

Environment tested

Harness (e.g. Claude Code, Cursor)	Harness version	Model	Model version/ID
Cursor cloud agent (this checkout)	Cursor cloud agent Linux VM, Oct 2026	Cursor Grok 4.6	cursor-grok-4.6-medium

Local: bash scripts/bump-version.sh --check — all seven files at 6.5.0. bash tests/skills/test-gemini-retirement.sh — PASS including ## v6.5.0 in release notes. npm run test:extended — PASS after installing host rsync (preflight exit 3 until that host tool was present; not a new plugin dependency). CI on this PR is the extended suite plus ShellCheck.

New harness support (required if this PR adds a new harness)

Not applicable. This PR does not add a harness.

Clean-session transcript for "Let's make a react todo list"

N/A — not a new harness integration.

Evaluation

• What was the initial prompt you (or your human partner) used to start the session that led to this change? 悟空非空也g asked to 发版 after cleanup, on current origin/main including docs: add GitHub issue and Projects collaboration rules #50 and docs: add Simplified Chinese product requirements document #51.
• How many eval sessions did you run AFTER making the change? Zero skill-behavior evals. This PR does not change skill or hook text.
• How did outcomes change compared to before the change? Version check and Gemini retirement notes gate now see 6.5.0; extended suite still passes. No agent-behavior claim.

Language-pack evidence

• Language and support status: n/a (no language pack change)
• Real failure generic methodology handled incorrectly: n/a
• Project markers and toolchain versions: n/a
• Official sources: n/a
• No-guidance repetitions and failures: n/a
• With-guidance repetitions and results: n/a
• Monorepo, unsupported-language, and docs-only controls: n/a
• Human reviewer familiar with the language: n/a
• Frameworks or third-party preferences introduced: none

Rigor

• If this is a skills change: I used wukong-code:writing-skills and completed adversarial pressure testing (paste results below)
• This change was tested adversarially, not just on the happy path
• I did not modify carefully-tuned content (Red Flags table, rationalizations, "human partner" language) without extensive evals showing the change is an improvement

Not a skills change. Adversarial here means: bump-version.sh --audit still reports unrelated fdir@6.5.0 strings in template lockfiles (same class of collision as any X.Y.Z that matches an npm package; those files stay out of .version-bump.json); Gemini retirement test requires the dated ## v6.5.0 heading; extended suite was run, not only --check.

Human review

• A human has reviewed the COMPLETE proposed diff before submission

悟空非空也g explicitly asked to 发版 this repo after cleanup, authorized a release PR targeting main, and already merged the only new product/docs commits in this cut (#50, #51) plus the Unreleased notes (#46). The diff in this PR is version metadata and notes for that already-reviewed main history. The human should still glance at RELEASE-NOTES.md wording on GitHub before relying on the GitHub Release object.