askill
executing-plans

executing-plansSafety 95Repository

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

0 stars
1.2k downloads
Updated 2/6/2026

Package Files

Loading files...
SKILL.md

Executing Plans

Overview

Load plan from beads, 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."

Storage: Plans are tracked in beads epics with markdown reference files for documentation.

The Process

The Process

Step 1: Load and Review Plan

Load from beads:

# Source helper functions
source skills/lib/beads-helper.sh

# Get the epic ID (from user or markdown file)
EPIC_ID="superpowers-abc123"

# View epic details
bd show "$EPIC_ID"

# List ready tasks (no blockers)
bd ready --parent "$EPIC_ID"

# List all tasks
bd list --parent "$EPIC_ID" --status open

Review process:

  1. Read epic description and task list from beads
  2. Review markdown reference file if available: docs/plans/YYYY-MM-DD-<topic>.md
  3. Review critically - identify any questions or concerns about the plan
  4. Check task dependencies with bd show <task-id>
  5. If concerns: Raise them with your human partner before starting
  6. If no concerns: Proceed with execution

No TodoWrite needed - beads tracks all state persistently.

Step 2: Execute Batch

Default: First 3 ready tasks

Get ready tasks:

# Get tasks with no blockers
bd ready --parent "$EPIC_ID" --format json | jq -r '.[0:3][] | .id'

For each task:

1. Mark as in_progress:

TASK_ID="superpowers-abc123.1"
bd update "$TASK_ID" --status in_progress

2. Get task details:

bd show "$TASK_ID"
# Review the description field for all steps

3. Follow each step exactly:

  • Plan has bite-sized steps in task description
  • Execute them one by one
  • Run verifications as specified

4. Mark as completed:

bd close "$TASK_ID"
# This automatically updates status to 'closed'
# Unblocks any dependent tasks

Between tasks: Let beads track the state. No manual status tracking needed.

Step 3: Report

When batch complete:

  • Show what was implemented (list completed task IDs)
  • Show verification output
  • Check remaining work:
    bd ready --parent "$EPIC_ID"
    bd list --parent "$EPIC_ID" --status open
    
  • Say: "Ready for feedback."

Step 4: Continue

Based on feedback:

  • Apply changes if needed
  • Update task status if reopening: bd update <task-id> --status open
  • Get next batch of ready tasks: bd ready --parent "$EPIC_ID"
  • Execute next batch
  • Repeat until complete

Check progress:

bd show "$EPIC_ID"  # Shows all children with status

Step 5: Complete Development

After all tasks complete and verified:

  • Check all tasks closed:
    bd list --parent "$EPIC_ID" --status open  # Should be empty
    bd show "$EPIC_ID"  # All children should show closed
    
  • 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
  • After merging/pushing, close the epic:
    bd close "$EPIC_ID"
    bd sync  # Sync beads to git
    

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.

Beads Integration

This skill reads implementation plans from beads epics and updates task status during execution.

Workflow:

  1. User provides epic ID (e.g., superpowers-abc123)
  2. Load epic and task list from beads
  3. Execute ready tasks in batches (no blockers)
  4. Update task status as you work (in_progress → closed)
  5. Beads automatically unblocks dependent tasks
  6. Continue until all tasks complete
  7. Close epic and sync to git

Key Commands:

  • Get epic: bd show <epic-id>
  • List ready tasks: bd ready --parent <epic-id>
  • List all tasks: bd list --parent <epic-id> --status <status>
  • Get task details: bd show <task-id>
  • Start task: bd update <task-id> --status in_progress
  • Complete task: bd close <task-id>
  • Close epic: bd close <epic-id>
  • Sync to git: bd sync

Task Status Flow:

open → in_progress → closed
      ↑ (if reopening)

Dependencies:

  • Beads tracks dependencies automatically
  • bd ready only shows tasks with no blockers
  • Closing a task unblocks its dependents
  • Check dependencies: bd show <task-id> (shows "depends on" section)

No TodoWrite:

  • Beads replaces TodoWrite entirely
  • All state persists in .beads/ directory
  • Synced to git via bd sync
  • Survives session restarts

Remember

  • Get epic ID from user or markdown reference file
  • Review plan critically first with bd show <epic-id>
  • Use bd ready to find unblocked tasks
  • Update beads status as you work (in_progress/closed)
  • Follow plan steps exactly from task descriptions
  • Don't skip verifications
  • Reference skills when plan says to
  • Between batches: report, sync beads, wait
  • Stop when blocked, don't guess
  • Close epic and sync when fully complete
  • 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

Install

Download ZIP
Requires askill CLI v1.0+

AI Quality Score

96/100Analyzed 2/12/2026

An exceptionally well-structured skill for executing implementation plans using the 'beads' tool. It provides clear, actionable steps, specific CLI commands, and robust error-handling guidelines.

95
98
75
98
98

Metadata

Licenseunknown
Version-
Updated2/6/2026
Publishermatthew-reed-holden

Tags

github-actionspromptingtesting