mi
Pass
Audited by Gen Agent Trust Hub on Aug 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill documentation and instructions are consistent with its stated purpose as a short alias for a larger skill set. No malicious code, credential harvesting, or unauthorized network activity was detected.
- [INDIRECT_PROMPT_INJECTION]: The skill identifies a surface where instructions are ingested from a secondary local file path.
- Ingestion points: The agent is instructed to load the file located at
.claude/skills/motivational-interviewing/SKILL.md. - Boundary markers: None; the agent is directed to follow the full content of the target file without specific delimiters.
- Capability inventory: The skill is configured with the
Readtool in its frontmatter to facilitate local file access. - Sanitization: No sanitization is performed on the content of the target file, which is expected for internal skill redirection.
Audit Metadata