stock-forecast-analysis
Pass
Audited by Gen Agent Trust Hub on Aug 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: In SKILL.md, the skill delegates technical analysis tasks to a sub-agent that executes the local script
scripts/ta.py. This script is part of the skill's technical seat and processes market data locally.\n- [EXTERNAL_DOWNLOADS]: SKILL.md and references/01-tipranks-methodology-and-sources.md specify a workflow that fetches information from official and well-known financial sources, including the SEC EDGAR system (data.sec.gov), TipRanks, and investor relations pages for earnings transcripts. These domains are trusted for financial data gathering.\n- [PROMPT_INJECTION]: The skill identifies a potential surface for indirect prompt injection due to the ingestion of untrusted external content like earnings transcripts and 8-K filings (SKILL.md). It mitigates this risk through a modular architecture using context-isolated sub-agents ('seats') to limit the agent's exposure to raw external text and requiring a 'skeptic gate' verification step for all findings.
Audit Metadata