← ClaudeAtlas

controlflow-planninglisted

Use when a repository task needs a strict ControlFlow-style implementation plan before coding, especially for small, medium, or large scope, cross-file edits, risky migrations, architectural uncertainty, or any work that should produce a saved Markdown plan artifact in plans/.
Smithbox-ai/ControlFlow · ★ 11 · AI & Automation · score 69
Install: claude install-skill Smithbox-ai/ControlFlow
# ControlFlow Planning ## Overview Create a durable, execution-ready plan. Keep the contract local: the saved artifact defines scope, risks, files, phases, validation, review route, and handoff without depending on VS Code agent mode or plugin-host-specific runtime contracts. Invoke this skill via `/controlflow-claude-code:controlflow-planning`. ## Local Contract - Read the repository before phase decomposition; keep verified facts separate from assumptions. - Use [references/plan-template.md](references/plan-template.md) as the plan contract, not a loose outline. - Apply [references/llm-behavior-guidelines.md](references/llm-behavior-guidelines.md) for generic assumption, simplicity, scope, and verification discipline. - Keep behavior portable: ask the user directly when a requirement changes scope, or record a bounded assumption when it does not. ## Workflow 1. Create a saved plan when the user asks for one or when work is SMALL, MEDIUM, or LARGE; skip it only for truly TRIVIAL work. 2. Save to `plans/<task-slug>-plan.md` unless the user names another path. 3. Clarify only when the answer changes file scope, user-visible behavior, architecture, or destructive-risk handling. 4. Map likely files, tests, commands, dependencies, and change boundaries before phase decomposition. 5. Read [references/complexity-tiers.md](references/complexity-tiers.md) and [references/semantic-risk-taxonomy.md](references/semantic-risk-taxonomy.md); assign one tier an