Skip to content
Get packs

Inline Plan Executor

Lets an agent carry out a written implementation plan itself, task by task, with a progress ledger and tests

Source Download skill
Repository
obra/superpowers
Stars
310,931
Licence
MIT
Files
1 file

What it does

When you already have a written implementation plan for a coding project and want the agent to build it in the current session, this skill tells the agent how to work through the plan step by step without spinning up separate helper agents for each task. It asks the agent to write tests first, check each step's output against the plan, keep a ledger file as a record of progress, and run one final review of the whole branch at the end. It also sets rules for when the agent should decide small issues itself and when it must stop and ask you.

Great for

  • Developers who have a fully written plan from a planning skill and want it implemented in one session to save cost
  • Teams whose agent tool has no subagent feature, so the plan must be run directly
  • Anyone who wants a traceable ledger of decisions made while the agent works unattended

Skip it if

  • Projects without a written plan, since the skill assumes one exists and is detailed (a vague plan makes the agent guess)
  • Long plans where the agent's context may be compacted, because later tasks get less attention; the file itself suggests a different approach in that case
  • Anything involving merges, pushes or publishing without your review, since the skill tells the agent to stop and ask for those but depends on the agent following that rule

Try asking

  • “Execute the implementation plan in docs/plans/payments.md inline in this session, following TDD and keeping a ledger”
  • “Run this plan here, no subagents available, and do a final review when all tasks are done”
  • “Pick up the plan where the ledger left off after the conversation was compacted”
Step-by-step guide: how to use Inline Plan ExecutorInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 1 original files, the MIT licence and an attribution note.

    executing-plans.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip executing-plans.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip executing-plans.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Give the agent a complete plan with exact expected outputs; the skill works best when the plan is fully specified
  • Choose inline execution deliberately and review the ledger afterwards, since decisions are recorded there rather than checked with you
  • Use the subagent-driven skill instead if you want a review after every task

What's inside

  • SKILL.md

Redistributed unmodified from obra/superpowers under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.