ingest-url
Pass
Audited by Gen Agent Trust Hub on Apr 25, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill includes a 'Content Trust Boundary' section. While this section contains keywords like 'ignore previous instructions', they are used defensively to instruct the agent to disregard any injection attempts found within the fetched web content. This is a security best practice for skills that process untrusted data.
- [COMMAND_EXECUTION]: The skill performs benign command execution using local tools such as
gitandgrepto detect project context and identify duplicate entries within the vault. These operations are scoped to the local environment and the intended functionality of managing a wiki. - [DATA_EXFILTRATION]: The skill uses
WebFetchto retrieve content from user-provided URLs. The instructions explicitly prohibit making any network requests beyond the target URL or exfiltrating data from the vault.
Audit Metadata