clawhub
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructions include installing the
clawhubCLI globally vianpmand fetching skills/data from the remote registry athttps://clawhub.com. These operations are central to the skill's primary function as a package manager. - [COMMAND_EXECUTION]: The skill provides numerous CLI command examples for authenticating, searching, installing, and updating skills on the local system.
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the download and installation of external agent skills, which represents a potential attack surface if remote content is malicious.
- Ingestion points: Remote skills downloaded via
clawhub installandclawhub updatefromhttps://clawhub.com(documented inSKILL.md). - Boundary markers: None specified; instructions do not include delimiters or warnings to ignore instructions embedded in downloaded content.
- Capability inventory: The skill utilizes
npmfor installation and theclawhubCLI for file management and network communication. - Sanitization: No evidence of automated sanitization, signature verification, or integrity checking of the downloaded skill content is present in the provided instructions.
Audit Metadata