Agent skill · Jesse Vincent (obra)

Verification Before Completion skill for Claude Code, Codex and Cursor

Requires running checks and showing their output before claiming work is done.

Communityname: verification-before-completion★ 305k (repo)pushed 2026-10-23MITSKILL.md on GitHubchecked 2026-11-07

What it does, in the publisher's words

Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always

The description field of the SKILL.md. The agent reads only this at startup and decides from it when to load the full skill.

Install Verification Before Completion

The skills CLI copies the skill into the folder of each agent you pick (Claude Code, Codex, Cursor, Copilot, Gemini CLI, OpenCode, Windsurf...). Add -g for all projects. Needs Node.js 22.20+. docs

npx skills add obra/superpowers --skill verification-before-completion

Works with

Claude Code, Claude.ai and Claude Desktop, Codex, GitHub Copilot, Cursor, Gemini CLI, OpenCode, Windsurf and other agents that read the open Agent Skills format.

What's in the SKILL.md

8 sections
  • Overview
  • The Iron Law
  • The Gate Function
  • Common Failures
  • Red Flags - STOP
  • Rationalization Prevention
  • Key Patterns
  • When To Apply

Facts

Skill name
verification-before-completion
Publisher
Jesse Vincent (obra) (community)
Repo
obra/superpowers
License
MIT
SKILL.md size
359 words, 116 lines
Repo stars
305k (as of 2026-10-24)
SKILL.md checked
2026-11-07

More skills from obra/superpowers

  • Brainstorming: Makes the agent clarify goals and explore options with you before building anything new.
  • Requesting Code Review: Has the agent get its finished work reviewed against requirements before merging.
  • Subagent-Driven Development: Executes a plan by handing each task to a fresh subagent with review between steps.
  • Systematic Debugging: A root-cause-first process for bugs and failing tests before any fix is attempted.
  • Test-Driven Development: Enforces a red-green-refactor loop: write a failing test first, then the code.
  • Using Git Worktrees: Sets up an isolated git worktree before feature work so the main checkout stays clean.
  • Writing Plans: Turns a spec into a detailed, step-by-step implementation plan before coding starts.
  • Writing Skills: Method for writing and testing new skills so they trigger and behave reliably.

Related testing & qa skills

FAQ

What does the verification-before-completion skill do?+

Requires running checks and showing their output before claiming work is done. In its own SKILL.md, Jesse Vincent (obra) describes it as: "Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always"

How do I install the verification-before-completion skill?+

Run npx skills add obra/superpowers --skill verification-before-completion and pick your agents, or use the tab for your agent above. In Claude Code the skill ends up in ~/.claude/skills/verification-before-completion/SKILL.md; in Codex and Cursor in .agents/skills/verification-before-completion/.

Does Verification Before Completion work in Codex, Cursor and Copilot, or only Claude?+

It's a standard SKILL.md, so any agent that reads the Agent Skills format can use it: Claude Code, the Claude apps, Codex, GitHub Copilot, Cursor, Gemini CLI, OpenCode, Windsurf and more.

Who maintains Verification Before Completion?+

Jesse Vincent (obra) (a community project), in the obra/superpowers repo on GitHub, last pushed 2026-10-23. License: MIT.

Read a skill's SKILL.md and scripts before you install it: skills aren't verified by any agent vendor. Claude skills marketplace · How to install skills · MCP servers