verification-before-completion
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of instructional markdown and metadata. It does not include any scripts, tools, or executable commands.
- [PROMPT_INJECTION]: The instructions reinforce behavioral guidelines for task completion and do not attempt to bypass safety filters or override core agent protocols.
- [EXTERNAL_DOWNLOADS]: The skill references an external GitHub repository for provenance and synchronization metadata, but no files are downloaded or executed at runtime.
- [INDIRECT_PROMPT_INJECTION]: While the skill discusses handling output from subagents, it explicitly instructs the agent to verify evidence (such as VCS diffs) rather than trusting self-reported success, which serves as a security best practice to mitigate indirect injection risks.
Audit Metadata