Claude-skill-registry implementing-a-plan
Loads plan, reviews critically, executes each batch, and reports for review between batches. Use when implementing a plan from a plan file.
install
source · Clone the upstream repo
git clone https://github.com/majiayu000/claude-skill-registry
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/majiayu000/claude-skill-registry "$T" && mkdir -p ~/.claude/skills && cp -r "$T/skills/data/implementing-a-plan" ~/.claude/skills/majiayu000-claude-skill-registry-implementing-a-plan && rm -rf "$T"
manifest:
skills/data/implementing-a-plan/SKILL.mdsource content
The Process
- Load plan and review critically
- Execute each batch with using a subagent in order
- Report for review between batches
- Incorporate feedback, if any
- Continue until all batches are done
Step 1: Load and Review Plan
- Read plan file
- Review critically. Identify any questions or concerns about the plan
- If concerns: Raise them with your human partner before starting
- If no concerns: Proceed with implementation
Step 2: Execute Batch
Dispatch fresh subagent
Task tool (general-purpose): description: "Implement Batch N" prompt: | You are implementing Batch N from [batch-file] created from [plan-file]. Read both the plan and the batch carefully. Your job is to: 1. Implement exactly what the batch file specifies 2. Verify implementation works 3. Commit your work 4. Report back Work from: [directory] Report: What you implemented, what you executed and their results, files changed, any issues
Step 3: Report
When batch is complete:
- Show what was implemented
- Show verification output
- Say: "Ready for feedback."
Step 4: Continue
- Incorporate and apply changes if needed
- Fill out the recap section of the next batch file
- Include any deviations from the plan that resulted from human feedback
- Include the agent output from the most recent batch
- Execute next batch
- Repeat until complete
When to Stop and Ask for Help
STOP executing immediately when:
- Hit a blocker mid-batch (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.
Core principle: Batch execution with checkpoints for architect review.
Remember
- Review plan critically first
- Follow plan steps exactly
- Don't skip verifications
- Reference skills when plan says to
- Between batches: just report and wait
- Stop when blocked, don't guess