analyse-comprehensive-crypto
Pass
Audited by Gen Agent Trust Hub on Aug 20, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides a shell script to initialize a temporary directory and launch Google Chrome with specific flags (
--remote-debugging-port=9222,--user-data-dir). This is a functional requirement to enable the TradingView MCP tools referenced in the skill. - [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow that ingests data from external sources (TradingView market data) and the natural language outputs of five subagents. This constitutes an attack surface where untrusted data could potentially influence the orchestrator's synthesis.
- Ingestion points: TradingView OHLCV data, indicator values, and results from
analyse-onchain,analyse-sentiment,analyse-macro,analyse-orderflow, andanalyse-narrativesubagents. - Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded prompts within the subagent responses.
- Capability inventory: The skill launches external processes (Chrome) and writes data to temporary files (
/tmp/chrome-cdp.log,/tmp/btc_analysis.png). - Sanitization: No explicit sanitization or validation of the market data or subagent output strings is implemented before synthesis.
Audit Metadata