documentation-audit
Warn
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: MEDIUMPROMPT_INJECTIONCOMMAND_EXECUTIONNO_CODE
Full Analysis
- [PROMPT_INJECTION] (MEDIUM): Indirect Prompt Injection Surface. The skill's core function is to ingest untrusted data from documentation files and use those claims to guide agent verification actions. * Ingestion points: Documentation files (e.g., CONFIGURATION.md) and codebase sources. * Boundary markers: Absent. There are no instructions or delimiters to isolate document text from agent commands. * Capability inventory: File system write access (audit reports) and CLI command verification. * Sanitization: Absent. Ingested content is processed without filtering for embedded instructions.
- [COMMAND_EXECUTION] (LOW): Potential command execution for CLI verification. The verification process for 'cli_command' claims includes checking if a 'Script supports it', which likely involves executing scripts with help flags or help commands.
Audit Metadata