pr
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill defines a natural language template for a pull request body, organizing information into Summary, Evidence, and Reversibility sections. It does not contain any shell commands, scripts, or network operations.
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest data from development session logs and test outputs. Ingestion points: 'OMC paper trail' and 'verify protocol' output (SKILL.md). Boundary markers: None. Capability inventory: No capabilities for command execution, network access, or file modification are defined. Sanitization: None. Because the skill lacks any dangerous capabilities, the processing of external logs does not present a security risk.
Audit Metadata