market-weather

Pass

Audited by Gen Agent Trust Hub on Feb 27, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local script macro-signal-tracker.py using parameters derived from user input or brand profiles (e.g., industry and time horizon). This introduces a potential command injection surface if the inputs are not strictly validated.
  • [PROMPT_INJECTION]: The skill is susceptible to Indirect Prompt Injection because it ingests and processes 'external signals' from cultural, platform, and regulatory sources which are not controlled by the user.
  • Ingestion points: Market signals retrieved by macro-signal-tracker.py in Step 2 and Step 3.
  • Boundary markers: No specific delimiters or instructions to ignore embedded commands are mentioned in the process.
  • Capability inventory: Accesses local filesystem (~/.claude-marketing/) and executes local Python scripts.
  • Sanitization: The process does not describe any sanitization or validation of the external signal data before review by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 27, 2026, 01:02 PM