signal-postmortem

Pass

Audited by Gen Agent Trust Hub on Apr 20, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADS
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches historical price data from the Financial Modeling Prep (FMP) API at financialmodelingprep.com. This is a well-known service for financial data.
  • [CREDENTIALS_UNSAFE]: The skill requires an API key (FMP_API_KEY) to fetch data. It correctly instructs the user to provide this via environment variables or command-line arguments rather than hardcoding secrets in the source code.
  • [DATA_EXFILTRATION]: Transmits the user-provided API key to the official service domain at financialmodelingprep.com to authenticate price data requests. This is expected behavior for using this service.
  • [PROMPT_INJECTION]: The skill creates a feedback loop that processes performance data to influence downstream components like the signal aggregator weights. This constitutes an indirect prompt injection surface.
  • Ingestion points: Reads signal records from JSON files provided via --signals-file or found in the signals directory.
  • Boundary markers: No specific delimiters or instructions to ignore embedded content are used when processing signal data.
  • Capability inventory: The skill has the ability to write JSON, YAML, and Markdown reports to the file system and make network GET requests to a financial API.
  • Sanitization: Signal data is parsed using standard JSON libraries without additional content sanitization.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 20, 2026, 09:30 AM