chore(main): release 1.2.0

[github-actions]

🤖 I have created a release beep boop

1.2.0 (2026-03-11)

Added

• add 6 CI checks — spell check, actionlint, frontmatter validation, llms.txt consistency, orphan detection, token budgets (bec24ab)
• add activation eval runner and updated test cases (08aac68)
• add CI integration for hook tests, banned phrases, and activation evals (4863054)
• add hook unit tests and banned phrase checker (55c6906)
• context-updater agent, token budget compliance, headless mode docs (#6) (a4b6159)

Fixed

• add robust stream-json parsing strategies for skill detection (0da56cc)
• correct spell check command in CLAUDE.md and allow HTML center in typos config (19af130)
• eval runner only flags ContextDocs skill activations as failures (d96899d), closes #5
• strengthen ai-context skill NL trigger descriptions (bd2503f), closes #4
• use unprefixed slash commands in activation evals (9028dcf), closes #3

Documentation

• optimise plugin for Anthropic best practices — trim skills, strengthen context files (b6ca593)
• update RESULTS.md with activation eval findings (30% pass rate) (71a6177)

---

This PR was generated with Release Please. See documentation.

[github-actions]

🤖 Created releases:

• v1.2.0

🌻