executing-plans
npx machina-cli add skill CodingCossack/agent-skills-library/executing-plans --openclawExecuting Plans
Overview
Load plan, review critically, execute tasks in batches, report for review between batches.
Core principle: Batch execution with checkpoints for architect review.
Announce at start: "I'm using the executing-plans skill to implement this plan."
Quick Reference
| Step | Action | Exit Condition |
|---|---|---|
| 1 | Load + Review | Concerns raised OR TodoWrite created |
| 2 | Execute Batch | 3 tasks complete + verified |
| 3 | Report | Show results, say "Ready for feedback" |
| 4 | Continue | Apply feedback, next batch |
| 5 | Complete | Use finishing-a-development-branch |
The Process
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: Create TodoWrite and proceed
Step 2: Execute Batch
Default: First 3 tasks
For each task:
- Mark as in_progress
- Follow each step exactly (plan has bite-sized steps)
- Run verifications as specified
- Mark as completed
Step 3: Report
When batch complete:
- Show what was implemented
- Show verification output
- Say: "Ready for feedback."
Step 4: Continue
Based on feedback:
- Apply changes if needed
- Execute next batch
- Repeat until complete
Step 5: 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:
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 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.
Common Mistakes
Executing without critical review
- Problem: Blindly following plan with gaps
- Fix: Step 1 explicitly requires raising concerns first
Batch size drift
- Problem: Doing 5-10 tasks without checkpoint
- Fix: Default is 3. Report after EVERY batch.
Guessing through blockers
- Problem: Making assumptions when stuck
- Fix: STOP immediately, ask for clarification
Skipping verification commands
- Problem: Mark complete without running tests
- Fix: Each task has verification - run it
Integration
Called by:
- writing-plans (Parallel Session choice)
Calls:
- finishing-a-development-branch (Step 5)
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
Source
git clone https://github.com/CodingCossack/agent-skills-library/blob/main/skills/executing-plans/SKILL.mdView on GitHub Overview
Executing-plans loads, critically reviews, and runs a saved implementation plan in batches. It emphasizes batch checkpoints for architect review and requires verifications between tasks. This approach helps catch gaps early and keeps stakeholders informed.
How This Skill Works
Begin by loading and critically reviewing the plan; raise concerns if any and create a TodoWrite if clear. Then execute in batches (default first 3 tasks), marking in_progress, following each step exactly, running verifications, and marking completed. After each batch, report results and say 'Ready for feedback,' then proceed or stop based on feedback, until all tasks are complete and verified, then run finishing-a-development-branch.
When to Use It
- Starting a saved implementation plan from a plan document.
- You need checkpointed execution with architect review between batches.
- Plans require explicit verifications for each task before completion.
- A partner updates the plan or you need to apply feedback and continue.
- All tasks are verified; you trigger the finishing-a-development-branch process to complete the work.
Quick Start
- Step 1: Load and critically review the plan; raise concerns if any.
- Step 2: Execute the first 3 tasks; mark in_progress, follow steps, run verifications, mark completed.
- Step 3: Report results, say 'Ready for feedback,' and await next batch (or completion).
Best Practices
- Read the plan file critically before starting and raise concerns if needed.
- Create TodoWrite only after confirming there are no blockers.
- Follow the plan steps exactly and perform the prescribed verifications.
- Execute in fixed-size batches (default 3) and report after every batch.
- Stop on blockers or unclear instructions; don't guess; move to feedback or clarifications.
Example Use Cases
- Implementing a feature per a saved plan, with batch verification.
- Applying a database migration plan with checkpoints.
- Deploying a microservice following an orchestration plan.
- Refactoring code with a step-by-step plan and test verifications.
- Config migrations executed in batches with QA sign-off.
