project-guide
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructions specify the execution of a Python script located at
scripts/project_guide.py. The source code for this script was not included in the provided skill files, preventing a full security assessment of its operations. - [INDIRECT_PROMPT_INJECTION]: The skill ingests content from the current workspace, including project repository files and technical descriptions. It lacks explicit boundary markers or sanitization mechanisms for this ingested data, making it potentially susceptible to indirect prompt injection if the project files contain malicious instructions.
- [INDIRECT_PROMPT_INJECTION]: Evidence Chain: 1. Ingestion points: Local project files and user materials (referenced in
SKILL.md). 2. Boundary markers: Absent. 3. Capability inventory: Command execution (scripts/project_guide.py) and file writing (documented inSKILL.md). 4. Sanitization: Semantic constraints on internal names are present, but structural sanitization is absent.
Audit Metadata