crawlbase
Pass
Audited by Gen Agent Trust Hub on Apr 21, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill guides the user to install the @membranehq/cli package from npm. This is a trusted resource as it is the official command-line tool for the vendor's platform.
- [COMMAND_EXECUTION]: The instructions include several shell commands using the membrane CLI to manage connections, discover actions, and execute API requests. These are standard operations for the integration.
- [DATA_EXFILTRATION]: The skill involves fetching content from external websites via Crawlbase, which introduces an indirect prompt injection surface. This is the primary function of the web crawling skill and is documented for transparency.
- Ingestion points: Web data enters the context via actions such as crawl-url and get-stored-result in SKILL.md.
- Boundary markers: No specific instructions are provided to delineate untrusted content from system instructions.
- Capability inventory: The skill uses membrane action run and membrane request for its operations.
- Sanitization: No sanitization steps for the ingested data are specified in the setup instructions.
Audit Metadata