review-tokens-pr
Pass
Audited by Gen Agent Trust Hub on Jun 20, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute standard local development commands (
pnpm build,pnpm test,pnpm typecheck,pnpm lint) to verify the integrity of the code. These commands are consistent with the skill's primary purpose of reviewing a software repository. - [PROMPT_INJECTION]: The skill contains an attack surface for indirect prompt injection as it processes Pull Request data (untrusted external input) and has the capability to execute commands. However, given the primary purpose of the skill is PR review, this is a standard operational surface.
- Ingestion points: Pull Request files including
.tokens.jsonand.changeset/*.md(SKILL.md). - Boundary markers: No specific delimiters or "ignore instructions" warnings are defined for the ingested content.
- Capability inventory: Shell command execution via
pnpm(SKILL.md). - Sanitization: No explicit sanitization or validation of the external content before command execution is described, relying on the built-in behavior of the build tools.
- [DATA_EXFILTRATION]: No sensitive file access or unauthorized network operations were identified. The skill focuses on local verification and standard PR feedback loops.
- [REMOTE_CODE_EXECUTION]: No patterns of downloading and executing remote scripts from untrusted sources were detected.
Audit Metadata