@6522/tangtrades

Pass

Audited by Gen Agent Trust Hub on Aug 6, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references the LISA_QUANT framework at github.com/tangtrades/lisa_quant for use in strategy development and backtesting. This repository is associated with the skill author's research methodology.
  • [DATA_EXFILTRATION]: Instructions include deploying interactive dashboards to Vercel and pushing code to GitHub repositories as part of the branded deliverable workflow.
  • [COMMAND_EXECUTION]: The agent is instructed to use headless Chromium for generating PDF reports and to write custom scripts for backtesting quantitative strategies when standard tools are inadequate.
  • [PROMPT_INJECTION]: The skill ingests data from external sources such as web searches, Twitter, and protocol documentation, creating a surface for indirect prompt injection. Ingestion points include web_search, web_fetch, twitter, and lunarcrush. Boundary markers include requirements to cite sources inline and cross-reference data across independent sources. Capability inventory includes trading tools for platforms like Hyperliquid and Binance, along with network-enabled delivery tools for GitHub and Vercel. Sanitization of external content is not explicitly mandated before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 6, 2026, 08:34 PM
Security Audit — agent-trust-hub — @6522/tangtrades