api-docs-generator
Warn
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION] (MEDIUM): The skill utilizes shell commands including
npx,pdoc, andmake. The use ofmakeis a security concern as it can execute arbitrary commands defined within a project'sMakefile. - [EXTERNAL_DOWNLOADS] (LOW): The skill uses
npxto dynamically download and execute Node.js packages (swagger-jsdoc,jsdoc) from the public npm registry at runtime, which introduces dependency on external registry integrity. - [PROMPT_INJECTION] (MEDIUM): The skill has an Indirect Prompt Injection surface (Category 8) because it parses comments in source code files. Evidence Chain: 1. Ingestion: Reads
routes/*.js,src/, and Python package files. 2. Boundary markers: Absent; no delimiters or instructions exist to prevent the agent from obeying instructions found in comments. 3. Capability inventory: Shell command execution and file output. 4. Sanitization: None; the content is processed directly by documentation tools.
Audit Metadata