@349/lighter-dex
Pass
Audited by Gen Agent Trust Hub on Mar 18, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCREDENTIALS_UNSAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires the installation of the
lighter-sdkPython package. It performs network operations tomainnet.zklighter.elliot.aiandtestnet.zklighter.elliot.aito retrieve market data and execute trades. These endpoints represent the official infrastructure for the Lighter DEX service. - [CREDENTIALS_UNSAFE]: The skill manages highly sensitive Lighter API private keys. It includes helper scripts like
generate_keys.pyandfind_account_index.pyto facilitate setup. The documentation consistently follows security best practices by advising users to store these credentials in environment variables rather than hardcoding them in scripts. - [COMMAND_EXECUTION]: The skill contains several executable Python scripts in the
scripts/directory designed for user-initiated account configuration, market lookup, and risk calculation. These scripts operate as intended for the primary purpose of the skill. - [INDIRECT_PROMPT_INJECTION]:
- Ingestion points: Market data, funding rates, and account balances are ingested from the Lighter DEX API in
tools.py. - Boundary markers: The skill does not use explicit boundary markers when passing API-retrieved data to the agent's context.
- Capability inventory: The skill possesses significant capabilities, including placing market/limit orders, modifying leverage, and canceling orders in
tools.py. - Sanitization: The skill sanitizes incoming API data by casting values to specific numeric types (floats and integers) before returning them to the agent.
Audit Metadata