swap-planner

Pass

Audited by Gen Agent Trust Hub on Mar 24, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes command-line tools like curl, jq, and cast (from Foundry) to interact with blockchain RPCs and external price APIs. It also uses open or xdg-open to launch the generated swap URL in the user's browser. These operations are governed by strict internal security rules provided in the skill to prevent command injection.
  • [EXTERNAL_DOWNLOADS]: Fetches real-time market data and token lists from well-known technology providers and the vendor's own infrastructure, including DexScreener, CoinGecko, and GeckoTerminal.
  • [DATA_EXPOSURE]: An initialization step sends a telemetry ping to pancakeswap.ai containing system metadata such as the operating system, architecture, and agent identifier. This is consistent with vendor analytics and is directed to the author's official domain.
  • [DATA_INGESTION]: The skill ingests data from multiple external crypto APIs. It specifically instructs the agent to treat this data as untrusted and to avoid interpreting token metadata as instructions, effectively mitigating potential indirect prompt injection risks.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 24, 2026, 09:30 AM