Agent skill
compound-workflows
Use Plan → Work → Review → Compound to compound skills and maintain project context.
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/data/compound-workflows
Metadata
Additional technical details for this skill
- tags
- workflow,compounding
- version
- 1
- created at
- 2026-01-27T17:03:09.731823+00:00
- updated at
- 2026-01-27T17:03:09.731823+00:00
SKILL.md
Purpose
This repository uses a loop:
- Plan: turn an idea into a ticketed plan (loom ticket), informed by recalled notes (loom memory).
- Work: execute in an isolated git worktree (loom workspace), updating ticket status.
- Review: run a multi-angle review before merging.
- Compound: extract reusable patterns into skills (procedural memory) + store memos for future planning.
The point is not vibes. The point is reusable procedure.
Commands
/workflows:plan <idea>
- Recall relevant memos for planning:
compound_memory_recall(query="<idea>", command="workflows:plan")
- Create/organize tickets:
compound_ticket(argv=["init"])(if needed)compound_ticket(argv=["create", "..."])compound_ticket(argv=["dep-add", "<id>", "<dep-id>"])(optional)
- Output a plan with:
- ticket IDs
- sequencing / dependencies
- acceptance criteria and tests
- risk list
/workflows:work <ticket-id>
- Fetch ticket, set status to
in_progress. - Create/ensure a worktree:
- Branch convention:
ticket-<id>-<slug> compound_workspace(argv=["repo", "worktree", "add", "<branch>"])
- Branch convention:
- Do the work in that worktree.
- Update ticket as you go (
add-note,update --status).
/workflows:review <ticket-id>
- Run fast checks (lint/tests if applicable).
- Do a review pass with multiple lenses:
- correctness & maintainability
- security & foot-guns
- performance/regressions
- docs & ergonomics
- Update the ticket with required follow-ups.
/workflows:compound <ticket-id>
- Write memory notes (loom memory) that future planning can recall.
- Propose skill changes as a CompoundSpec v1 JSON object.
- Call
compound_apply(spec_json=...)to make it real:- create/update skills under
.opencode/skills/ - update AI-managed blocks in AGENTS/PROJECT/ROADMAP
- append an agent-optimized CHANGELOG entry
- sync derived indexes
- create/update skills under
Operational defaults
- Keep skills small, scoped, and action-oriented.
- Prefer updating an existing skill over creating a near-duplicate.
- A skill should be applicable in at least 2 future contexts.
Manual notes
This section is preserved when the skill is updated. Put human notes, caveats, and exceptions here.
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
Didn't find tool you were looking for?