Dev & Engineering verificationtestingbuild-validationcode-qualitygit-workflow

Verification Before Completion

Require fresh evidence before claiming work is complete, fixed, or passing.

FollowSkills review · FSRS-2.0
Use with care
48/ 100 5-point scale 2.4 / 5
1 2 3 4 5 6
1Trust12 / 25 · 2.4/5

The skill requires fresh verification evidence, complete output review, exit-code checking, and truthful status reporting, reducing false completion claims; however, it mandates running the full command without defining command permissions, sensitive-data handling, destructive-operation confirmation, rollback, or data-flow disclosure, so points are deducted.

2Reliability8 / 20 · 2.0/5

The workflow is internally clear: identify the proving command, run it, read output, check failures, and report actual status; however, it lacks diagnostic guidance for missing dependencies, command failures, environment differences, or inability to proceed, and key paths were not executed during this static review, so the calibration limit reduces the score.

3Adaptability9 / 15 · 3.0/5

Triggers and scenarios are fairly clear, covering tests, builds, fixes, commits, PRs, task completion, and agent delegation; however, the rule covering any positive wording is overly broad, with limited non-fit boundaries, platform guidance, and Chinese-language support, so points are deducted.

4Convention10 / 15 · 3.3/5

The skill has a name, description, overview, core rule, failure tables, and examples; repository context supplies MIT licensing, a version, and some contribution and maintenance signals. The skill itself lacks installation/dependency notes, changelog/version guidance, troubleshooting, explicit ownership, and an update path, so points are deducted.

5Effectiveness6 / 15 · 2.0/5

The central goal—preventing unsupported completion claims—is directly addressed with actionable rules and examples; however, there is no independent enforcement mechanism or skill-specific verification evidence, so effectiveness depends on agent compliance and static review cannot confirm representative usable outcomes.

6Verifiability3 / 10 · 1.5/5

The source is auditable and specifies required evidence, counterexamples, and verification steps; however, there are no dedicated tests for this skill, third-party execution records, or corroborating evidence, leaving coverage limited and warranting a low score.

Evidence confidence:Low Reviewed Jul 19, 2026 Reviewed revision d884ae04edeb
The upstream repository has new commits since this review. The score still applies to the reviewed revision shown and may not cover the latest changes.
Before you use it
  • “Run the full command” may create an overly broad execution scope; define command provenance, permission boundaries, sensitive-output handling, and confirmation and recovery for destructive operations before use.
  • The skill largely relies on the agent following textual instructions and provides no executable enforcement or dedicated tests.
  • The rule covering any positive statement may over-trigger; define exclusions for ordinary discussion, planning, and conditional statements.
Review evidence [1][2][3][4][5][6][7]
See the full review method →

What does this skill do, and when should you use it?

Verification Before Completion is a development workflow skill in the Superpowers collection. It requires an agent to obtain fresh verification evidence before claiming that work is complete, a bug is fixed, or tests or builds pass. The agent must identify the command that proves the claim, run the full command, read its output and exit code, count failures, and compare the evidence with the intended statement. The skill does not prescribe a specific test, build, or lint command.

Before making a completion or success claim, it requires the agent to identify the proving command, execute the complete command, read the full output, inspect the exit code, count failures, and determine whether the evidence supports the claim. It also applies before commits, pull requests, task transitions, and agent delegation; unsupported claims must be replaced with the actual evidenced status.

  1. A developer finishes a change and needs to confirm the test suite actually passes before reporting completion.
  2. A developer fixes a bug and needs to rerun the original symptom test to establish that the fix works.
  3. A developer is preparing a build, commit, or pull request and needs evidence from the complete relevant command.
  4. A team uses subagents and needs to independently inspect their reports, version-control changes, and verification results.

What are this skill's strengths and limitations?

Pros
  • Enforces fresh, complete verification evidence.
  • Covers tests, builds, bug fixes, requirement checks, commits, pull requests, and delegated work.
  • Explicitly requires reporting the actual status when evidence does not support the claim.
Limitations
  • Does not automatically choose a project-specific test or build command.
  • Provides no standalone installation procedure or executable script in the supplied material.
  • Its effectiveness depends on the agent identifying and running the correct complete command.

How do you install this skill?

This skill is one of 14 skills bundled in the obra/superpowers repository. The README does not document a standalone installation procedure for this skill; install the Superpowers collection through the relevant harness. For Codex CLI, open /plugins, search for superpowers, and select Install Plugin. In Codex App, install Superpowers from the Plugins sidebar under Coding.

How do you use this skill?

Apply it immediately before reporting completion, a fix, or a passing result. Example trigger: Run the complete verification command that proves the test result, read the full output and exit code, then report the evidenced status. The source does not define one universal command; the agent must choose the command appropriate to the claim and project.

FAQ

Does it automatically run tests?
No. It requires the agent to identify and run the complete command that proves the relevant claim, but it does not prescribe a universal test command.
Are previous test results sufficient?
No. The skill requires fresh verification evidence from the current work, so an earlier run is not sufficient for a new claim.
What happens when verification fails?
The agent must state the actual evidenced status instead of claiming success.
What work does it cover?
It covers development claims involving tests, builds, bug fixes, requirement checks, commits, pull requests, task transitions, and delegated agent work.

More skills from this repository

All from obra/superpowers

Dev & Engineering

Executing Plans

Turn a written implementation plan into verified, checkpointed development work.

Dev & Engineering

Subagent-Driven Development

Execute independent implementation tasks with fresh agents and review gates.

Dev & Engineering

Development Branch Finisher

Safely close completed development work after tests pass, with guided merge, PR, retention, or discard choices.

Dev & Engineering

Systematic Debugging

Find the root cause before changing code, reducing trial-and-error rework.

Dev & Engineering

Continuous Code Review

Dispatch focused reviews after tasks, major features, and before merges to catch problems early.

Dev & Engineering

Rigorous Code Review Reception

Verify review feedback before changing code, then act on technical evidence.

Dev & Engineering

Isolated Git Workspace

Creates an isolated workspace for feature work, then prepares the project and verifies its clean test baseline.

Dev & Engineering

Writing Plans

Turn multi-step software requirements into executable, testable implementation plans.

Dev & Engineering

Test-Driven Development

Prove expected behavior with a failing test before writing implementation code.

Dev & Engineering

Design Before Code

Clarify intent, compare approaches, and secure an approved design before implementation begins.

Dev & Engineering

Test-Driven Skill Authoring

Create, revise, and verify Agent Skills with a test-driven documentation workflow.

Dev & Engineering

Superpowers Skill Bootstrap

Ensures applicable skills are discovered and invoked before every response or action.

Dev & Engineering

Parallel Agent Dispatch

Delegate independent engineering problems to focused agents and investigate them concurrently.

Related skills