update implementation plan based on recent work

[justinmclean]

Summary

Type of change

• Skill change (.claude/skills/<name>/) — eval fixtures updated below
• Tool / bridge contract (tools/<system>/*.md)
• Python package (tools/*/ with pyproject.toml)
• Groovy reference impl
• Cross-cutting (RFC, AGENTS.md, sandbox, privacy-LLM)
• Documentation (docs/, README.md, CONTRIBUTING.md)
• Project template (projects/_template/)
• CI / dev loop (prek, workflows, validators)
• Other:

Test plan

• prek run --all-files passes
• For Python packages touched: uv run pytest / ruff check / mypy passes
• For Groovy bridges touched: command-line invocation tested end-to-end
• For skill changes: eval suite passes for the affected skill
  (PYTHONPATH=tools/skill-evals/src python3 -m skill_evals.runner tools/skill-evals/evals/<skill>/)
• For skill behaviour changes: a new or updated eval fixture is included in this PR
  (a regression test for the bug fixed / the behaviour added — see CONTRIBUTING.md)
• Other:

[choo121600]

LGTM — clean planning-doc refresh: the plan is rewritten to track only open gaps, the In-flight table now carries verifiable Implemented by commits, the new maintainer-education.md spec is well-formed, every cross-reference resolves, and CI is green. Two nits below, neither gating.

I checked the things a docs/planning change can actually get wrong:

• No dangling links introduced — every path the new spec points at exists on main: docs/index.md, docs/rfcs/RFC-AI-0004.md, tools/privacy-llm/pii.md, tools/skill-evals/, plus the MISSION/PRINCIPLES anchors.
• Spec matches the sibling convention — specs/maintainer-education.md carries the SPDX header and the same frontmatter shape as specs/skill-reconciler.md / specs/reviewer-routing.md (title / status / kind / mode / source / acceptance), and the standard section set (What it does / Where it lives / Behaviour & contract / Out of scope / Acceptance criteria / Validation / Known gaps).
• Internally consistent — work items 2–6 correctly realize "one landing page plus one per MISSION-named piece" (the four pieces: pattern catalogue, your-first-skill, eval-driven development, workshops); item 7 is properly flagged as an epic that must be decomposed before building; and the new overview.md row (proposed — release-blocking per PRINCIPLE 18) is consistent with the spec's status: proposed.

Smaller observations

• PR body ## Summary is empty. For a change that deletes 384 lines and re-scopes the whole work queue, a one-line "why now" — the merge train shipped the old items 1–22, so the plan refocuses on mail-privacy README compliance and the release-blocking education stream — would help future readers of the plan's history. Nit, not blocking.
• .typos.toml — extend-ignore-re = ["[0-9a-f]{7,40}"]. Correctly scoped to backtick-wrapped hex and pairs well with the new Implemented by commit column. Just noting it will also silence a genuine ≥7-char all-hex token that happens to be backtick-wrapped (e.g. a stray `deadbeef`-style word) — vanishingly rare in prose, so fine as-is.

---

This review was drafted by an AI-assisted tool and confirmed by an Apache Magpie maintainer. The maintainer approving this PR has read the findings and signed off. If something feels off, please reply on the PR and a maintainer will follow up.

More on how Apache Magpie handles contributions: CONTRIBUTING.md.

[choo121600]

looks good :)

[potiuk]

LGTM — docs-only plan refresh. The plan correctly stops mirroring shipped state (deferring that to specs/, code, and git history) and re-centres on the maintainer-education stream; the new .typos.toml ignore rule matches the in-flight table's backtick-hash column exactly, and the new specs/maintainer-education.md carries its SPDX header and stays placeholder-clean. CI green. (Second maintainer sign-off — already approved by @choo121600.)