SkillRepo
Sign InSign Up
Catalog
obra/executing-plans

obra

executing-plans

Use when you have a written implementation plan to execute in a separate session with review checkpoints

global
0installs0uses—~604
v1.0Saved May 2, 2026

Executing Plans

Overview

Load plan, review critically, execute all tasks, report when complete.

Announce at start: "I'm using the executing-plans skill to implement this plan."

Note: Tell your human partner that Superpowers works much better with access to subagents. The quality of its work will be significantly higher if run on a platform with subagent support (such as Claude Code or Codex). If subagents are available, use superpowers:subagent-driven-development instead of this skill.

The Process

Step 1: Load and Review Plan

  1. Read plan file
  2. Review critically - identify any questions or concerns about the plan
  3. If concerns: Raise them with your human partner before starting
  4. If no concerns: Create TodoWrite and proceed

Step 2: Execute Tasks

For each task:

  1. Mark as in_progress
  2. Follow each step exactly (plan has bite-sized steps)
  3. Run verifications as specified
  4. Mark as completed

Step 3: Complete Development

After all tasks complete and verified:

  • Announce: "I'm using the finishing-a-development-branch skill to complete this work."
  • REQUIRED SUB-SKILL: Use superpowers:finishing-a-development-branch
  • Follow that skill to verify tests, present options, execute choice

When to Stop and Ask for Help

STOP executing immediately when:

  • Hit a blocker (missing dependency, test fails, instruction unclear)
  • Plan has critical gaps preventing starting
  • You don't understand an instruction
  • Verification fails repeatedly

Ask for clarification rather than guessing.

When to Revisit Earlier Steps

Return to Review (Step 1) when:

  • Partner updates the plan based on your feedback
  • Fundamental approach needs rethinking

Don't force through blockers - stop and ask.

Remember

  • Review plan critically first
  • Follow plan steps exactly
  • Don't skip verifications
  • Reference skills when plan says to
  • Stop when blocked, don't guess
  • Never start implementation on main/master branch without explicit user consent

Integration

Required workflow skills:

  • superpowers:using-git-worktrees - REQUIRED: Set up isolated workspace before starting
  • superpowers:writing-plans - Creates the plan this skill executes
  • superpowers:finishing-a-development-branch - Complete development after all tasks
Files1
1 files · 1.0 KB

Select a file to preview

Overall Score

78/100

Grade

B

Good

Safety

88

Quality

72

Clarity

82

Completeness

65

Summary

A meta-skill that guides an AI agent through executing a previously written implementation plan in a structured, checkpoint-driven manner. It emphasizes critical review before execution, task-by-task verification, and explicit escalation to human partners when blockers are encountered. The skill requires integration with git-worktrees, plan-writing, and branch-finishing workflows.

Detected Capabilities

Plan file reading and critical reviewTask progress tracking (in_progress, completed states)Step-by-step instruction following with verificationBlocker detection and escalation to human partnersIntegration with related workflow skills (git-worktrees, finishing-a-development-branch)Git branch safety checks (prevents main/master commits without consent)

Trigger Keywords

Phrases that MCP clients use to match this skill to user intent.

execute planimplement planfollow implementationcheckpoint reviewmulti-step development

Risk Signals

INFO

Requires subskill integration (finishing-a-development-branch) as mandatory workflow step

Step 3: Complete Development section
INFO

Agent autonomy is explicitly scoped — agent must stop and ask for help on blockers rather than guess

When to Stop and Ask for Help section
INFO

References external plan file without specifying format or validation — agent must assume plan is well-formed

Step 1: Load and Review Plan
INFO

Defers critical safety decision (starting work on main/master) to explicit user consent

Remember section, final bullet

Use Cases

  • Execute a detailed development plan across multiple sessions with checkpoints
  • Implement features with built-in verification steps and progress tracking
  • Work through multi-step technical tasks with explicit decision gates
  • Coordinate multi-session development with clear handoff points

Quality Notes

  • POSITIVE: Clear decision trees for when to stop and ask for help — reduces risk of agents plowing through problems
  • POSITIVE: Explicit announcement pattern ('I'm using the executing-plans skill...') improves transparency and auditability
  • POSITIVE: Integration points with other skills are documented (git-worktrees, finishing-a-development-branch) and marked as REQUIRED
  • POSITIVE: Safety boundaries are explicit — won't start work on protected branches without consent, won't skip verifications
  • POSITIVE: Handles plan updates gracefully (Step 1 revisit path) without forcing agent to restart mid-execution
  • NEGATIVE: Plan file format is not specified — agent has no documented schema to validate against (relies on human-provided plan quality)
  • NEGATIVE: 'TodoWrite' is referenced in Step 1 but never defined or explained — unclear what artifact this creates
  • NEGATIVE: Verification steps are deferred to plan itself ('Run verifications as specified') — skill doesn't provide guidance on what constitutes valid verification
  • NEGATIVE: No guidance on handling partial completion or partial plan failures (what if task N succeeds but task N+1 fails?)
  • NEGATIVE: Integration with superpowers:writing-plans is noted in workflow but not explained in this skill's body — assumes user already has a plan
Model: claude-haiku-4-5-20251001Analyzed: May 2, 2026

Reviews

Add this skill to your library to leave a review.

No reviews yet

Be the first to share your experience.

Add obra/executing-plans to your library

Get Started Free

Command Palette

Search for a command to run...