rate-skill
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided SKILL.md files to perform audits. This ingestion of untrusted data creates a potential surface for indirect prompt injection, where instructions within the audited file could attempt to influence the agent's grading or logic.\n
- Ingestion points: The agent reads the entire content of the target SKILL.md file as specified in the workflow.\n
- Boundary markers: None identified; the agent parses the full content of the file without explicit delimiters for untrusted sections.\n
- Capability inventory: The skill utilizes Read, Glob, Grep, and Bash(python3 *) tools.\n
- Sanitization: There is no explicit sanitization or filtering of the file content before processing.\n- [COMMAND_EXECUTION]: The skill uses the Bash tool, restricted to python3 *, to execute a local scoring script (measure.py) located in a relative path. The script is used for rubric calculation and weighted scoring, which is a standard part of the skill's auditing functionality.
Audit Metadata