trade

Pass

Audited by Gen Agent Trust Hub on Aug 8, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests and processes untrusted market data from the web.\n
  • Ingestion points: The skill uses web_search and web_extract to fetch news, sentiment, analyst ratings, and transcripts (SKILL.md).\n
  • Boundary markers: No protective delimiters or instructions to ignore embedded commands are present in the orchestrator instructions.\n
  • Capability inventory: The agent has access to terminal, execute_code, and task delegation tools, which could be exploited if malicious instructions are processed.\n
  • Sanitization: No evidence of data sanitization or validation of the scraped content is specified.\n- [COMMAND_EXECUTION]: The skill executes a local Python script (generate_trade_pdf.py) located in the skill's installation directory (~/.hermes/skills/trade/scripts/). This execution is part of the skill's intended functionality for report generation.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 8, 2026, 02:43 PM
Security Audit — agent-trust-hub — trade