skills/executing-plans/SKILL.md
Use when you have a written implementation plan to execute in a separate session with review checkpoints
npx skillsauth add schlenks/superpowers-bd executing-plansInstall 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.
Load beads epic, review critically, execute tasks in dependency-aware batches, report for review between batches.
Announce at start: "I'm using the executing-plans skill to implement this plan."
REQUIRED BACKGROUND: Read superpowers-bd:beads SKILL.md first (bd CLI usage, permission avoidance, dependency management).
Prerequisites: Beads epic exists (via plan2beads), dependencies set (bd blocked shows expected blockers).
references/codex-execution-checkpoints.md for native progress tracking, batch checkpoints, user feedback, and verification evidence. Use update_plan for execution phases and beads for durable issue state.bd show <epic-id>, check bd ready and bd blocked. Review critically -- raise concerns before starting.
Batch = all currently ready issues (no blockers). For each:
bd update <id> --status=in_progress then bd show <id>git diff --cached --stat -- any file >50 lines changed -> apply the appropriate rule-of-five variant: Skill(superpowers-bd:rule-of-five-code) for code, Skill(superpowers-bd:rule-of-five-tests) for test files. Complete all 5 passes before proceedingbd close <id> -- unblocks dependent issuesFile conflicts: serialize issues touching same file. See references/batch-execution-detail.md.
Create checkpoint task after each batch. Show: implemented work, verification output, rule-of-five variant passes, newly unblocked issues (bd ready). Say: "Ready for feedback." See references/report-and-feedback.md.
Create blocked feedback + execution tasks to enforce checkpoint. Do NOT proceed until human feedback received. See references/report-and-feedback.md.
All epic issues closed -> verify with bd show <epic-id> -> hand off to superpowers-bd:finishing-a-development-branch. See references/completion-protocol.md.
Batch 1: All tasks with no blockers (bd ready)
Close completed -> check bd ready again
Batch 2: Newly unblocked tasks
Close completed -> check bd ready again
See references/batching-example.md for full walkthrough.
bd ready shows ALL epics. Filter to current epic: bd show <epic-id> for child IDs, only work on issues that are both ready AND children of your epic.
STOP when: blocker mid-batch, no ready issues but open issues remain, unclear instructions, repeated verification failures. Check bd blocked for deadlocks. See references/epic-scoping-and-deadlock.md.
Ask for clarification rather than guessing.
bd ready before each batch, close issues promptly to unblock dependentsrule-of-five-code for code, rule-of-five-tests for tests)references/batch-execution-detail.md: Full Step 2 procedure with file conflict handling and rule-of-five-code triggerreferences/codex-execution-checkpoints.md: Codex-native batch execution, checkpoint, and feedback flowreferences/report-and-feedback.md: Steps 3+4 TaskCreate blocks and feedback enforcement flowreferences/batching-example.md: Dependency-aware batching walkthrough with bd ready/blocked at each stagereferences/epic-scoping-and-deadlock.md: Epic scoping, deadlock detection, bd failures, when to revisitreferences/completion-protocol.md: Step 5 completion handoff to finishing-a-development-branchtools
Use when converting a Superpowers-BD implementation plan or Shortcut story into a beads epic with dependency-aware child tasks
development
Use when the user asks for /cr-style review of local changes, commits, a branch diff, or a GitHub PR outside subagent-driven development
development
Use when you have a spec or requirements for a multi-step task, before touching code
tools
Use when starting any conversation - establishes how to find and use skills, requiring Skill tool invocation before ANY response including clarifying questions