nws-active-alerts
Fail
Audited by Gen Agent Trust Hub on Aug 18, 2026
Risk Level: HIGHCREDENTIALS_UNSAFEREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [REMOTE_CODE_EXECUTION]: The documentation instructs users to download a JavaScript client (
call.mjs) from a remote server (https://fittings.sh/skill/call.mjs) usingcurland execute it locally usingnode. - [CREDENTIALS_UNSAFE]: The skill utilizes a blockchain private key for payment signing. The instructions direct the user to store this sensitive credential in the
FITTINGS_PRIVATE_KEYenvironment variable, where it is accessed in plaintext by thecall.mjsscript. - [COMMAND_EXECUTION]: The
call.mjsscript uses thenode:fsandnode:cryptomodules to read its own source code from the file system and generate a SHA-256 hash for integrity verification against a remote endpoint. - [EXTERNAL_DOWNLOADS]: The skill downloads and installs several external dependencies at runtime, including
@x402/fetch,@x402/evm, andviemto facilitate cryptocurrency transactions. - [INDIRECT_PROMPT_INJECTION]: The skill processes live weather alert data from the National Weather Service API through a third-party gateway (
fittings.sh). The ingested data is interpolated into the agent's context, creating a surface for potential injection if the remote data is malformed or malicious.
Recommendations
- AI detected serious security threats
Audit Metadata