npm-search
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from the public npm registry (package names and descriptions) which are untrusted external sources. Attackers can publish packages with malicious instructions in the metadata to influence the agent's behavior.
- Ingestion points: Results from
bash scripts/npmsearch(skill.md). - Boundary markers: No delimiters or specific instructions to ignore embedded content are present.
- Capability inventory: Shell command execution via bash (skill.md).
- Sanitization: The skill does not describe any sanitization or filtering of search results before they are processed by the agent.
- [COMMAND_EXECUTION]: The skill invokes shell commands using user-provided queries as arguments (
bash scripts/npmsearch "<query>"). This pattern relies on external scripts (scripts/npmsearch) to handle input safely.
Audit Metadata