ln-614-docs-fact-checker
Pass
Audited by Gen Agent Trust Hub on May 12, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill parses and processes project documentation files that may contain untrusted data. This establishes an indirect prompt injection surface where malicious instructions placed in documentation could influence the agent during the fact-checking process.
- Ingestion points: Project markdown files (.md) discovered through Glob and Read tools.
- Boundary markers: There are no explicit delimiters or specific instructions for the agent to treat processed documentation content as data distinct from the skill's own operational rules.
- Capability inventory: The skill has access to the Bash, Read, Grep, and Glob tools, enabling arbitrary command execution and file system access.
- Sanitization: Documentation content is processed directly for claim extraction without validation or sanitization of embedded instructions.
- [COMMAND_EXECUTION]: The skill uses the Bash tool to verify factual claims in documentation. This includes running system commands such as 'ls' to check file existence, 'grep' to verify code content or configuration keys, and 'sed' to inspect specific line references. These operations are intended for documentation auditing and are documented in the skill's workflow.
Audit Metadata