web-reader
Pass
Audited by Gen Agent Trust Hub on Jul 14, 2026
Risk Level: SAFEDATA_EXFILTRATIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill transmits user-provided URLs to the Jina Reader service (r.jina.ai) for processing. The documentation provides clear guidance to avoid sending URLs containing credentials or private tokens to this external service.
- [COMMAND_EXECUTION]: The skill identifies a fallback mechanism using bash and requests for scenarios where the primary tool is blocked.
- [PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by ingesting untrusted data from the web.
- Ingestion points: Arbitrary external URLs are processed into Markdown content.
- Boundary markers: The instructions do not define delimiters to separate untrusted content from the agent's instructions.
- Capability inventory: The agent has access to shell execution (bash) and network tools.
- Sanitization: No sanitization of the retrieved Markdown is performed before it is added to the agent's context.
Audit Metadata