← ClaudeAtlas

controlflow-orchestrationlisted

Use when executing a multi-step repository task in phases, especially when progress should be tracked with gates, approvals, retries, localized replanning, or optional sub-agent delegation instead of one long monolithic run.
Smithbox-ai/ControlFlow · ★ 11 · AI & Automation · score 69
Install: claude install-skill Smithbox-ai/ControlFlow
# ControlFlow Orchestration ## Overview Execute an approved plan with explicit state, disciplined gates, bounded retries, and local evidence. Keep ControlFlow's orchestration discipline without assuming VS Code agent runtime contracts. Invoke this skill via `/controlflow-claude-code:controlflow-orchestration`. ## Local Contract - Start from a saved plan, preferably `plans/<task-slug>-plan.md`; create one with `/controlflow-claude-code:controlflow-planning` if needed. - Confirm strict plan review before executing non-trivial work. - Track current phase, last verification, blocker, and next action. - Apply [../controlflow-planning/references/llm-behavior-guidelines.md](../controlflow-planning/references/llm-behavior-guidelines.md) for generic scope and verification discipline. ## Workflow 1. Confirm plan path, current phase, target files, validation command, and approval needs. 2. For non-trivial work, run or verify review routing first: - SMALL: plan-audit - MEDIUM/LARGE: plan-audit + assumption-verifier - LARGE: add executability-verifier - unresolved HIGH risk: include assumption-verifier 3. If review blocks, revise the plan before implementation. 4. Use wave execution only when write ownership is clearly disjoint; otherwise serialize phases. 5. Keep immediate blocking work in the current conversation context. Use Claude Code sub-agents only when the task benefit is clear and the user wants delegation. 6. Use [references/failure-taxonomy.md](refe