Verification Before Completion
PassAudited by ClawScan on May 1, 2026.
Overview
This instruction-only skill coherently tells the agent to verify work before claiming success, with no code, credentials, persistence, or hidden install behavior.
This appears safe for its purpose. Before installing, understand that it may make the agent run fresh verification commands such as tests, builds, linters, or diffs before claiming success; guide the agent if those commands are costly or have side effects in your project.
Findings (1)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
The agent may run project verification commands before saying work is complete, which could take time or trigger normal build/test side effects.
The skill directs the agent to run verification commands before making success claims. This is purpose-aligned for a verification skill, but it means users should expect test, build, lint, or similar local commands to be executed.
RUN: Execute the FULL command (fresh, complete)
Use this skill in repositories where running tests, builds, or linters is acceptable, and review or guide the verification command if the project has expensive or side-effecting checks.