Skip to content

chore(agents): add MQ dual skills bundle#100

Merged
MCamner merged 3 commits into
mainfrom
chore/mq-dual-skills
Jun 22, 2026
Merged

chore(agents): add MQ dual skills bundle#100
MCamner merged 3 commits into
mainfrom
chore/mq-dual-skills

Conversation

@MCamner

@MCamner MCamner commented Jun 22, 2026

Copy link
Copy Markdown
Owner

What

Installs the MQ dual Claude/Codex skills bundle for mq-agent and adds short routing snippets to the committed agent surfaces.

  • 5 skills under both surfaces (byte-identical): .claude/skills/ (Claude Code) and .agents/skills/ (Codex) — mq-writing-plans, mq-worktree-safe, mq-subagent-driven-development, mq-interactive-command-design, mq-secrets-public-safe.
  • Short ## MQ Skills routing section added to CLAUDE.md and AGENTS.md (routing only — no skill bodies duplicated).

Why these skills

Per the MQ repo matrix, mq-agent owns planning + workflow routing, so it gets the planning/worktree/execution/interactive/public-safe set.

Notes

  • Skills come from mq-skillpack-dual-v2, sanitized: the public-safe / threat-model scan lines no longer embed private identifiers — they use portable placeholders (<your-username> / <your-hostname> / <location>).
  • CLAUDE.md / AGENTS.md are generated from mqobsidian templates; the hand-added routing block is overwritten on next regeneration. Durable fix = add the block to the mqobsidian generator templates (follow-up).
  • No runtime/application code changed.

Verification

  • .claude/skills == .agents/skills (dual layout identical)
  • Public-safe scan clean (only the skills' own scan patterns match — documentation)

🤖 Generated with Claude Code

@MCamner MCamner merged commit 5d10cf8 into main Jun 22, 2026
5 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant