project-personality
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes common system utilities and git commands (git, wc, grep, sed, sort, uniq) to extract repository metrics. These operations are local and perform non-destructive analysis consistent with the skill's stated purpose.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external repository sources (file lists, commit history, and code comments containing TODOs), which represents an indirect prompt injection surface. This is a common characteristic of code analysis tools.\n
- Ingestion points: File lists, commit logs, and content matched via git grep in the target repository (SKILL.md).\n
- Boundary markers: Not explicitly implemented in the agent's processing instructions for the raw data.\n
- Capability inventory: Shell command execution for metadata and history retrieval.\n
- Sanitization: None identified; the agent evaluates raw output from the analysis commands.
Audit Metadata