feat: multi-agent templates with mandatory commit, configurable push and QA cadence#1
feat: multi-agent templates with mandatory commit, configurable push and QA cadence#1GenieRobot merged 9 commits intomainfrom
Conversation
…thub#1971) * fix: unify hyphenated skills and migrate legacy kimi dotted dirs * fix: preserve legacy kimi dotted preset skill overrides * fix: migrate kimi legacy dotted skills without ai-skills flag * fix: harden kimi migration and cache hook init options * fix: apply kimi preset skill overrides without ai-skills flag * fix: keep sequential branch numbering beyond 999 * test: align kimi scaffold skill path with hyphen naming * chore: align hook typing and preset skill comment * fix: restore AGENT_SKILLS_DIR_OVERRIDES compatibility export * refactor: remove AGENT_SKILLS_DIR_OVERRIDES and update callers * fix(ps1): support sequential branch numbers above 999 * fix: resolve preset skill placeholders for skills agents * Fix legacy kimi migration safety and preset skill dir checks * Harden TOML rendering and consolidate preset skill restore parsing * Fix PowerShell overflow and hook message fallback for empty invocations * Restore preset skills from extensions * Refine preset skill restore helpers * Harden skill path and preset checks * Guard non-dict init options * Avoid deleting unmanaged preset skill dirs * Unify extension skill naming with hooks * Harden extension native skill registration * Normalize preset skill titles
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
…ithub#1989) Bumps [DavidAnson/markdownlint-cli2-action](https://github.com/davidanson/markdownlint-cli2-action) from 19 to 23. - [Release notes](https://github.com/davidanson/markdownlint-cli2-action/releases) - [Commits](DavidAnson/markdownlint-cli2-action@v19...v23) --- updated-dependencies: - dependency-name: DavidAnson/markdownlint-cli2-action dependency-version: '23' dependency-type: direct:production update-type: version-update:semver-major ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Bumps [actions/deploy-pages](https://github.com/actions/deploy-pages) from 4 to 5. - [Release notes](https://github.com/actions/deploy-pages/releases) - [Commits](actions/deploy-pages@v4...v5) --- updated-dependencies: - dependency-name: actions/deploy-pages dependency-version: '5' dependency-type: direct:production update-type: version-update:semver-major ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
- Extension ID: plan-review-gate - Version: 1.0.0 - Author: luno - Catalog entries sorted alphabetically by ID - README table row inserted alphabetically by name Co-authored-by: Ed Harrod <your-real-email@luno.com> Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Adds the onboard extension (v2.1.0) — contextual onboarding and progressive growth for developers new to spec-kit projects. - 7 commands: start, explain, trail, quiz, badge, mentor, team - 3 hooks: after-implement, before-implement, after-explain - Repository: https://github.com/dmux/spec-kit-onboard
Adds 7 extensions forming the MAQA (Multi-Agent & Quality Assurance) suite to catalog.community.json in correct alphabetical order (after 'learn', before 'onboard') and to the README community extensions table: - maqa — coordinator/feature/QA workflow, board auto-detection - maqa-azure-devops — Azure DevOps Boards integration - maqa-ci — CI/CD gate (GitHub Actions/CircleCI/GitLab/Bitbucket) - maqa-github-projects — GitHub Projects v2 integration - maqa-jira — Jira integration - maqa-linear — Linear integration - maqa-trello — Trello integration All entries placed alphabetically. maqa v0.1.3 bumped to reflect multi-board auto-detection added in this release.
Top-level updated_at was 00:00:00Z while plan-review-gate entries had 08:22:30Z, making metadata inconsistent for freshness consumers. Updated to 2026-03-27T08:22:30Z (>= all entry timestamps).
…ush and QA cadence
- Add templates/agents/{coordinator,feature,qa}.md — language-agnostic
versions of the three-agent workflow (coordinator → feature → QA)
- feature agent: commit is non-negotiable before returning result;
optional git push gated on auto_push setting
- coordinator: reads auto_push and qa_cadence from .specify/init-options.json;
qa_cadence=per_feature (default) spawns QA after each feature agent,
qa_cadence=batch_end waits until all batch features are done to save credits
- qa agent: git commit verification as step 0 prevents work loss on worktree deletion
- Add --auto-push/--no-auto-push and --qa-cadence CLI flags to specify init
- Save both settings to .specify/init-options.json with safe defaults
(auto_push=false, qa_cadence=per_feature)
- Fix scaffold_from_core_pack to recursively copy templates/ subdirectories
so offline scaffolding matches release ZIP output (parity test coverage)
- Bundle templates/agents/ into wheel via pyproject.toml force-include
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
There was a problem hiding this comment.
Cursor Bugbot has reviewed your changes and found 1 potential issue.
Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.
| ( | ||
| migrated_legacy_kimi_skills, | ||
| removed_legacy_kimi_skills, | ||
| ) = _migrate_legacy_kimi_dotted_skills(skills_dir) |
There was a problem hiding this comment.
Codex dotted skill names not migrated unlike Kimi
Medium Severity
The _compute_output_name change converts dots to hyphens in skill names for all skill-backed agents (e.g. speckit-test.plan → speckit-test-plan), but _migrate_legacy_kimi_dotted_skills only runs for Kimi. Codex users with existing extension skills containing internal dots (like speckit-myext.cmd) will end up with orphaned old directories alongside newly created hyphenated ones. Preset unregistration may also fail to match old skill names stored in the registry against the new naming convention.


Summary
templates/agents/{coordinator,feature,qa}.md— language-agnostic three-agent workflow templatesgit pushgated onauto_pushsetting (prevents work loss when worktrees are deleted)auto_pushandqa_cadencefrom.specify/init-options.json; supportsqa_cadence=per_feature(QA after every feature) orqa_cadence=batch_end(QA once when whole batch is done, saves credits)--auto-push/--no-auto-pushand--qa-cadenceflags tospecify init; saved to.specify/init-options.jsonwith safe defaults (auto_push=false,qa_cadence=per_feature)scaffold_from_core_packto recursively copytemplates/subdirectories (previously only copied top-level files, breaking parity with release ZIP for any template subdirectory)templates/agents/into wheel viapyproject.tomlforce-includeTest plan
tests/test_multi_agent_config.pycovering defaults, flag values, and validationtest_parity_bundled_vs_release_script) pass for all 28 agents🤖 Generated with Claude Code
Note
Medium Risk
Updates core project initialization, skill naming/migration, and extension/preset skill registration logic; regressions could affect existing projects’ agent skills and hook invocations across multiple AI assistants.
Overview
Adds bundled multi-agent agent templates (
coordinator,feature,qa) and persists new init options (--auto-push,--qa-cadence) into.specify/init-options.jsonfor downstream tooling.Unifies skill directory naming to hyphenated
speckit-*for Kimi and Codex (including extension/preset-generated skills), adds a one-time migration for legacy Kimi dotted directories, and updates hook messaging to emit the correct invocation style ($speckit-*for Codex skills,/skill:speckit-*for Kimi).Improves robustness of skill/override rendering: recursive bundling of
templates/(including subdirs), safer path rewriting for scripts/templates/memory references, TOML prompt rendering that handles embedded triple quotes, and sequential branch numbering that supports 4+ digit prefixes (with PowerShelllongparsing).Written by Cursor Bugbot for commit be151b6. This will update automatically on new commits. Configure here.