tools
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to "Read and follow" instructions from the file
.agents/workflows/tools.mdstep by step. This creates an attack surface where instructions contained in that file could override or influence agent behavior without explicit sanitization or boundary markers. - Ingestion points:
.agents/workflows/tools.md(referenced inSKILL.md) - Boundary markers: None specified in the instructions.
- Capability inventory: The skill directs the agent to follow steps within the target file, potentially granting the agent capability to perform any action described in that external source.
- Sanitization: No sanitization or validation of the file content is performed before processing.
Audit Metadata