verification-before-completion
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists entirely of instructional markdown text and a standard MIT license. It does not contain executable scripts, network operations, or access to sensitive file paths. Its primary function is to constrain agent behavior to ensure work is verified before status claims are made.
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read and verify output from external tools (e.g., test suites, linters). While tool output is technically an ingestion point for untrusted data (Category 8c), the skill itself does not provide an exploitable surface beyond standard agent operation and serves as a defense-in-depth measure by requiring evidence rather than trusting assertions.
Audit Metadata