fittings-fdic-bank-lookup
Pass
Audited by Gen Agent Trust Hub on Aug 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTIONCREDENTIALS_UNSAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches a JavaScript execution client (call.mjs) from the vendor's official domain at fittings.sh.
- [REMOTE_CODE_EXECUTION]: The skill executes the downloaded call.mjs script using the node runtime to perform API lookups and handle crypto-based payment settlement.
- [COMMAND_EXECUTION]: The skill instructs the agent to invoke shell commands including curl for downloading assets and node for script execution.
- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface where user-controlled arguments are interpolated into a shell command.
- Ingestion points: The 'name' and 'limit' parameters are provided by the user/agent at runtime in SKILL.md.
- Boundary markers: No boundary markers or delimiters are used to isolate user input from the command structure.
- Capability inventory: The skill utilizes curl and node for local execution and network communication.
- Sanitization: The skill does not define any sanitization or validation logic to prevent shell metacharacters from being injected into the execution string.
- [CREDENTIALS_UNSAFE]: The skill requires a blockchain private key for USDC payments on the Base network, passed through the FITTINGS_PRIVATE_KEY environment variable. The documentation acknowledges that this key is stored in plaintext within the agent's environment, posing a risk if the environment is compromised.
Audit Metadata