aegis-authz-access-control
Pass
Audited by Gen Agent Trust Hub on Aug 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection due to its core function of processing untrusted external codebases.
- Ingestion points: The agent is instructed to read and analyze various source code files, route definitions, and configuration files (e.g., urls.py, controllers/, Program.cs) from the project under audit, as specified in the 'Where To Look' and 'Patterns & Signatures' sections of SKILL.md.
- Boundary markers: The instructions lack guidance on using delimiters or protective prompting techniques to isolate untrusted file content from the agent's primary instructions.
- Capability inventory: The skill directs the agent to utilize search utilities (rg) and network tools (curl) to perform its audit and validation tasks, providing a path for potential exploits to leverage these capabilities if a malicious injection is successful.
- Sanitization: No validation or sanitization steps are defined for the data read from external files prior to its use in the agent's decision-making process.
Audit Metadata