npm-publishing
Pass
Audited by Gen Agent Trust Hub on Oct 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by processing external repository data while possessing sensitive capabilities.
- Ingestion points: The skill reads local project files using
cat package.jsonandls -la, and fetches external registry data vianpm viewandnpm access. - Boundary markers: There are no explicit instructions or delimiters provided to help the agent distinguish between the content of repository files and its own system instructions.
- Capability inventory: The skill uses the
run_commandtool, providing the agent with shell access, and performs sensitive operations likenpm publishandnpm trustwhich utilize user credentials or OIDC tokens. - Sanitization: The skill lacks logic to sanitize or validate the values retrieved from
package.json(such as scripts or version strings) before they are used in command construction or decision-making.
Audit Metadata