web-scraping
Pass
Audited by Gen Agent Trust Hub on Jun 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses
curlto interact with the vendor's API andpython3to parse local JSON credentials. It also references thenpx gooseworks logincommand for authentication. - [DATA_EXFILTRATION]: The skill reads API credentials from
~/.gooseworks/credentials.json. This access is necessary for authenticating with the vendor's official API (api.gooseworks.ai) and follows standard practices for the Gooseworks ecosystem. - [PROMPT_INJECTION]: As a tool designed to ingest content from arbitrary URLs, it is subject to indirect prompt injection (Category 8).
- Ingestion points: Multiple scraping, crawling, and browser automation endpoints ingest data from external websites.
- Boundary markers: The instructions do not specify any delimiters or warnings to the agent regarding embedded instructions in scraped data.
- Capability inventory: The agent can perform network requests via
curl, execute shell commands, and read local files. - Sanitization: No specific sanitization or filtering logic is mentioned for the processed web content.
Audit Metadata