executing-plans

An agent skill by obra, from obra/superpowers. Tags: api, developer-tools, web.

What it does

Use when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available

Install

With the skills CLI, which installs into Claude Code, Codex, Cursor and other agents:

npx skills add obra/superpowers --skill executing-plans

Or copy the skill folder into Claude Code's skills directory by hand (~/.claude/skills for every project, or .claude/skills inside one):

git clone --depth 1 https://github.com/obra/superpowers
cp -r superpowers/skills/executing-plans ~/.claude/skills/executing-plans

Safety box score

Grade C, moderate risk: 26 of 100 risk points, rated 2026-10-20. The grade rates what the skill and its scripts can reach on the machine of whoever installs it, across eight categories from shell execution to secrets access. How the score works.

This skill is a workflow for executing a plan inline in a git worktree: it runs task-start/task-done/review helpers, writes a ledger, commits work, and performs a final review. The main risks are broad autonomous branch manipulation, scratch-workspace destruction via cleanup, and dependence on helper scripts whose behavior is not visible in this file.

Category levels, from 0 (none) to 3 (high)
Category	Level	Why
Code & shell execution	2 (moderate)	It explicitly tells the agent to run shell helpers and task commands such as `scripts/task-start`, `scripts/task-done`, and `review-package`, so it can execute project commands inside the session.
Network access	0 (none)	SKILL.md does not describe downloading from or sending data to remote hosts.
Filesystem reach	1 (low)	It creates and writes a per-plan workspace/ledger under `.superpowers/sdd/...` and says `git clean -fdx` "will destroy the workspace," so it can erase scratch files in that area.
Secrets & credentials	0 (none)	There is no instruction to read, store, or request tokens, passwords, key files, or other credentials.
Privilege & persistence	0 (none)	The skill does not request sudo, global installation, services, shell startup edits, or other privilege changes.
Instruction hijack surface	1 (low)	It pressures the agent to skip mid-task human check-ins (`"Do not pause to check in with your human partner between tasks"`) and continue through conflicts unless one of four stop conditions applies.
Irreversible actions	2 (moderate)	The workflow is explicitly autonomous for the whole plan (`"Execute all tasks from the plan without stopping"`) and includes commits and end-of-branch cleanup, though it stops short of push/deploy behavior.
Transparency	1 (low)	Core actions are delegated to helper scripts and sub-skill files (`scripts/task-start`, `scripts/task-done`, `review-package`, `../subagent-driven-development/scripts/sdd-workspace`) whose contents are not shown here, so their exact side effects are not fully auditable from this file.

Before installing

Inspect the helper scripts (`scripts/task-start`, `scripts/task-done`, `review-package`, `sdd-workspace`) to confirm they stay inside the worktree and do not use network or privileged actions.
Be comfortable with a workflow that may continue without mid-task human check-ins and may delete the scratch workspace at the end.
Try it once in a disposable branch/worktree to observe the ledger and cleanup behavior before using it on important work.

Rated 2026-10-20 by an automated review of 1 file from the skill's repository.

Source

Repository: obra/superpowers (all skills from this repository)
Path: skills/executing-plans/SKILL.md
Branch: main
Updated: 2026-09-19

finishing-a-development-branch — Use when implementation is complete, all tests pass, and you need to decide how to integrate the work
receiving-code-review — Use when receiving code review feedback, before implementing suggestions.
requesting-code-review — Use when completing tasks, implementing major features, or before merging to verify work meets requirements
systematic-debugging — Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes
test-driven-development — Use when implementing any feature or bugfix, before writing implementation code
using-git-worktrees — Use when starting feature work that needs isolation from current workspace or before executing implementation plans - ensures an isolated workspace exists via.