verify-patterns
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to read and analyze external files including agent workflow definitions, tool implementations, and prompt templates (e.g., graph.py, tools.py, prompts.py, system.md). These files represent untrusted ingestion points where an attacker could embed malicious instructions to subvert the analysis or the agent's behavior.
- Ingestion points: The skill reads files listed in Step 2, such as graph.py, tools.py, state.py, prompts.py, prompts/*.md, and system.md.
- Boundary markers: No specific delimiters or 'ignore embedded instructions' warnings are provided for the agent when processing these files.
- Capability inventory: The skill is intended to perform read and process operations across the local file system to generate reports.
- Sanitization: There is no evidence of sanitization or filtering of the content read from these external files before the agent processes them for pattern checks.
Audit Metadata