strategy-discovery
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external local backtest artifacts and manifest files, creating a surface for indirect prompt injection.
- Ingestion points: JSON manifest files and directory-based backtest artifacts including CSV and JSON files (e.g., trades.csv, equity.csv, state.json).
- Boundary markers: No specific boundary delimiters or 'ignore instructions' warnings are documented for tool outputs containing data from processed artifacts.
- Capability inventory: The skill is capable of reading files from the local filesystem and writing to a local cache database.
- Sanitization: Ingested data is subject to hard-gated validation checks that verify process success, file existence, and numeric metric integrity before being accepted.
Audit Metadata