analyse-smartmoney-form4

Pass

Audited by Gen Agent Trust Hub on Aug 20, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches public financial data from authoritative and well-known sources including the SEC EDGAR system (sec.gov) and OpenInsider (openinsider.com).
  • [COMMAND_EXECUTION]: Executes a local Python helper script (watch.py) provided within the skill package to manage a ledger of recommendations and prevent duplicate alerts. The script uses standard Python libraries and performs local file I/O within the user's cache directory.
  • [DATA_EXPOSURE]: Maintains a local JSONL ledger in the .cache directory to store historical ticker recommendations for deduplication purposes. The skill does not access sensitive system files, environment variables containing credentials, or private user data.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external SEC filings and third-party financial screeners. While this presents an ingestion surface where an adversary could theoretically attempt to embed instructions in public filing fields (e.g., in a company name or insider title), the skill's execution environment is constrained to scoring and logging financial transactions.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 20, 2026, 02:21 PM
Security Audit — agent-trust-hub — analyse-smartmoney-form4