verification-before-merge
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from GitHub issues and pull requests, creating a surface for indirect prompt injection attacks.\n
- Ingestion points: Fetches PR reviews, review threads, and issue bodies using
gh pr viewandgh issue viewcommands inSKILL.md.\n - Boundary markers: The instructions lack specific delimiters or system instructions to treat external text as data rather than instructions, potentially allowing embedded commands to be followed.\n
- Capability inventory: The agent has significant capabilities including
gh pr merge,git push --force-with-lease, and executing local build scripts viapnpm.\n - Sanitization: No sanitization or filtering of the retrieved GitHub metadata is performed before the agent evaluates it for gate completion.\n- [COMMAND_EXECUTION]: The skill invokes several local development tools and environment managers to perform its tasks.\n
- Evidence: Executes
pnpm test,pnpm migrate,docker-compose ps, and variousgitandghcommands throughoutSKILL.mdto validate and merge code.
Audit Metadata