clawhub

Warn

Audited by Gen Agent Trust Hub on Mar 30, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONCREDENTIALS_UNSAFEDATA_EXFILTRATION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Installs the clawhub command-line utility globally via the npm package manager.
  • [REMOTE_CODE_EXECUTION]: Provides functionality to download and install external agent skills from the remote clawhub.com registry. This allows the agent to dynamically acquire and execute unvetted code from a third-party source.
  • [COMMAND_EXECUTION]: Executes various shell commands for managing skills, including npm install, clawhub install, and clawhub update.
  • [CREDENTIALS_UNSAFE]: References authentication mechanisms (clawhub login) that involve handling access tokens or credentials for the remote registry.
  • [DATA_EXFILTRATION]: Facilitates the transmission of local skill directories and associated metadata to an external server via the publish command.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 30, 2026, 12:15 AM
Security Audit — agent-trust-hub — clawhub