skills/decompose/SKILL.md
Break a clear goal, plan, or project into independently executable pieces with dependencies and verification criteria. Use when work is too large for one coherent implementation slice or when delegation needs clean ownership boundaries.
npx skillsauth add sofer/.agents decomposeInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Turn a known goal into executable pieces that can be sequenced, delegated, verified, and recombined.
Use this after the problem and constraints are clear. If they are not clear, use problem-statement or constrain first.
routine: clear work, suitable for delegationjudgement: needs human decision pointsdiscovery: approach is not known yetexternal: blocked on outside input or approval## Execution plan
### Phase 1: [Name]
#### Task 1.1: [Name]
- Type: routine | judgement | discovery | external
- What: [One sentence]
- Depends on: [Dependencies or none]
- Produces: [Output]
- Done when: [Verifiable criteria]
- Owner: [If known]
## Critical path
[Tasks where delay delays the whole project.]
## Parallel opportunities
[Tasks that can be done independently.]
## Checkpoints
[Where the user should review or decide.]
tools
Check whether Claude and Codex have equivalent access to shared agent resources, skills, hooks, plugins, MCP servers, permissions, startup behaviour, and provider-specific adapter config. Use when comparing agent environments, debugging missing capabilities after restart, or deciding whether to symlink a resource or configure a runtime.
testing
Record substantive skill use in an append-only local log. Use after choosing or invoking a non-system skill for real work, when a skill is inspected but not used, or when a skill fails to apply. Do not use for routine system skills or incidental file reads.
testing
Turn a vague or underspecified request into a self-contained problem statement. Use when the user has a rough idea, when a request would fail if handed directly to an agent, or before non-trivial work that needs shared understanding.
data-ai
Append a one-line learning to ~/.agents/learning-log.md. Use when the user types /learning, or when something genuinely worth remembering surfaced during work and the user confirms it should be captured.