Bundle the TSL API index and markdown references under the skill, route TSL docs to use it, and update the codegen index generator to maintain the bundled data.
Move generated builtin and dotnet function docs into the upgraded taxonomy, remove legacy route pages, and regenerate function_index.tsv from markdown headings.
Group playbook.toml-driven behavior tests under a single config-actions file and split parser/template contract coverage into focused files.
Remove obsolete PowerShell audit scripts and tracked validation reports; write template validation reports under reports/templates.
Use plan-status as the only machine state source for plan queueing, claiming, and finishing.
Route record-plan through queue insertion and update templates/prompts to match the single-state model.
BREAKING CHANGE: playbook.py -record-spec and main_loop.py record are removed.
- Add TSF file naming constraint to syntax/02_core_model.md
- Clarify semicolon rules: syntax facts vs style preferences
- Separate control flow end semicolon rules (syntax allows both)
- Add function body semicolon requirements to syntax/05_functions_and_calls.md
- Move style preferences to code_style.md (control flow end semicolons)
- Remove cross-references from syntax docs to maintain independence
- Enhance Gitea workflow emoji for better CI output readability
- Fix CI test path from tests/ to test/
- Organize agent test results under test/agent/result/ directory
- Add complete Chinese translation of test cases (test_cases_zh.md)
- Clean up .gitignore to use unified test/agent/result/ directory
- Remove obsolete agent test artifacts (REPORTS_LOCATION.md, old results)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Validate required Plan Meta before claims, record claim ownership, and persist verification evidence on completion.
Document conditional skill triggers without adding them to automatic main-loop dispatch.
Claim the first executable plan in plan-file order after resuming any in-progress plan, preserve skipped as its own workflow phase, and document the required post-done commit gate in the agent rules.
restore the missing typing.Optional import in scripts/playbook.py
so older python runtimes can import the CLI module successfully.
add a regression test for typing annotation imports and register it
in the tests README.
move the repository skills source from `codex/skills` to `skills`.
update vendor/install_skills paths, thirdparty sync pipeline, docs, and regression tests, including deployment-route e2e coverage.
- Simplify docs/ section to grouped summary with pointer to docs/index.md
- Remove outdated performance metrics table
- Use cross-platform path examples throughout
- Make AGENT_RULES tool usage section platform-agnostic
- Auto-create AGENT_RULES.local.md on sync_rules (empty template)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
When CLAUDE.md lives in a subdirectory (e.g. .claude/CLAUDE.md), the
@AGENTS.md import must use ../AGENTS.md since Claude Code resolves @path
relative to the file's directory. Calculate depth automatically.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Add skill_link config option to create a symlink/junction from another
platform's skills dir to agents_home/skills/ (e.g. ~/.claude -> ~/.agents)
- On Windows, falls back to directory junction when symlink requires admin
- Add _create_skills_symlink() with _is_junction() helper for Windows
- Update playbook.toml.example with skill_link documentation
- Fix templates/README.md prompt example to be platform-agnostic
- Add 3 tests: symlink creation, idempotency, absence when unconfigured
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
When any sync action runs and the target project already has a CLAUDE.md,
automatically insert @AGENTS.md / @AGENT_RULES.md import block so
Claude Code picks up the same rules as Codex without manual setup.
- Add sync_claude_md() called at end of sync_agents_template()
- Uses <!-- playbook:claude:start/end --> block markers for idempotent updates
- Skips if no CLAUDE.md exists (Codex users unaffected)
- Skips if @AGENTS.md already present (manual setups respected)
- Add 4 tests covering: no file, append, update block, already-referenced
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Remove docs/tsl/syntax_book/ (superseded by docs/tsl/syntax/)
- Remove data/tsl_reference_catalog_source/ (catalog already generated
into docs/tsl/reference/catalog/, source no longer needed)
- Remove scripts/build_tsl_reference_catalog.ps1 (build script for
the above source)
- Remove test assertion that referenced the deleted data/ source index
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
remove obsolete main_language placeholders and language summary template text
limit generated .agents guidance to configured langs and normalize AGENT_RULES trailing newlines
document install_skills no_backup behavior and refresh tests for deployment links and sync flows
replace the old select/record progress flow with a single main_loop claim/finish CLI.
route template execution through main_loop only, remove the legacy plan_progress entry points, and update tests to enforce the new behavior.
normalize Windows path-like TOML config values, regenerate .agents/index.md on sync, keep the SKILLS.md superpowers section route-only, and ignore generated Python cache dirs.
add regression coverage for load_config() and the standards sync behavior touched by these fixes.
csh
