npm-publish
Pass
Audited by Gen Agent Trust Hub on Mar 29, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted data from
git logand project files (CHANGELOG.md,README.md) to generate release documentation. This presents an indirect prompt injection surface where malicious commit messages or file content could attempt to influence agent behavior. \n - Ingestion points: Git commit history via
git log,package.json,CHANGELOG.md, andREADME.md(detected in SKILL.md).\n - Boundary markers: Absent; the skill lacks delimiters or explicit instructions to ignore embedded commands in the ingested text.\n
- Capability inventory: Subprocess execution via
npm install,npm test,npx tsc, andnpm pack(detected in SKILL.md).\n - Sanitization: Absent; the skill formats external text directly into the changelog and documentation files.\n- [COMMAND_EXECUTION]: The skill executes several shell commands to manage dependencies and verify the package, including
npm install,npm test,npm audit,npx tsc, andnpm pack. These are routine operations for the skill's stated purpose of preparing a software release and are restricted to the local project environment.
Audit Metadata