writing-pr-bodies
Pass
Audited by Gen Agent Trust Hub on Nov 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest content from project briefs and incorporate it into PR bodies, which creates a surface for instructions in the brief to potentially influence the agent's actions.\n
- Ingestion points: Project briefs and source code (SKILL.md).\n
- Boundary markers: No explicit delimiters or boundary markers are defined to isolate untrusted content from the brief.\n
- Capability inventory: The skill requires the agent to execute shell commands such as
git diff,tsc, andeslintto generate report data (pr-body-template.md).\n - Sanitization: There is no evidence of sanitization or filtering for the ingested brief content.\n- [COMMAND_EXECUTION]: The skill directs the agent to use local development tools (
git diff --stat,tsc --noEmit,eslint) to verify changes and paste the resulting output into the PR template. These are standard and expected capabilities for a development-focused agent skill.
Audit Metadata