breakout-trade-planner
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from VCP screener JSON files to generate trade reports.\n
- Ingestion points: The
load_inputfunction inscripts/plan_breakout_trades.pyreads user-provided JSON files from the local file system.\n - Boundary markers: The skill does not implement explicit boundary markers or warnings within the generated markdown reports to distinguish between calculated data and raw input strings.\n
- Capability inventory: Analysis of the source code confirms the skill has no network capabilities, no subprocess execution, and no dynamic code evaluation. Its actions are restricted to reading input files and writing text/JSON reports to an output directory.\n
- Sanitization: The input is parsed using standard JSON libraries and validated against a schema, but text fields are not sanitized for potential natural language instructions that could influence the LLM when reading the generated report.
Audit Metadata