crawlberg-serving-the-api
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest data from external web sources via scraping and crawling endpoints.
- Ingestion points: Untrusted content enters the agent's context through URL-based scraping and crawling requests (e.g., via
/v1/scrapeand/v1/crawlendpoints described in SKILL.md). - Boundary markers: The instructions do not specify the use of delimiters or warnings to ignore instructions embedded within the scraped HTML or markdown data.
- Capability inventory: The skill utilizes the crawlberg binary to perform network fetches and process file data.
- Sanitization: No explicit sanitization or filtering steps are documented for the external content before it is processed by the agent.
- [COMMAND_EXECUTION]: The skill instructs the agent or user to execute shell commands to set up and manage the service.
- Evidence: Documentation includes the command
crawlberg serve --host 127.0.0.1 --port 3000to start the API server. - [EXTERNAL_DOWNLOADS]: The skill provides instructions for installing the necessary CLI tool from a package registry.
- Evidence: Recommends the command
cargo install crawlberg-cli --features apifor from-source builds.
Audit Metadata