clari-performance-tuning
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from the external Clari API, creating a surface for indirect prompt injection. 1. Ingestion points: The client.export_and_download method in SKILL.md. 2. Boundary markers: The Output section specifies that the report should use aggregates and redacted failure reasons, providing limits on data propagation. 3. Capability inventory: The skill utilizes file writing for caching and makes network requests via the Clari client. 4. Sanitization: The provided code does not explicitly sanitize the entries returned by the API before processing.
- [SAFE]: The skill follows security best practices by recommending that API credentials be stored outside of the source code and version control systems.
Audit Metadata