internal-links
Pass
Audited by Gen Agent Trust Hub on Oct 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data sources which could contain malicious instructions.
- Ingestion points: The skill reads user-supplied drafts, pastes of URL lists, and content from external
/sitemap.xmlfiles. - Boundary markers: There are no explicit instructions to the agent to treat these inputs as data only or to ignore embedded instructions.
- Capability inventory: The agent has the ability to perform network fetches (sitemap retrieval) and modify text drafts based on the input.
- Sanitization: The skill does not define any sanitization or validation logic for the ingested content.
- [COMMAND_EXECUTION]: The instructions direct the agent to perform network operations, specifically fetching a site's
/sitemap.xmlfile, which involves executing tools or shell-like commands to interact with external domains. - [DATA_EXFILTRATION]: While not explicitly malicious, the skill reads a local context file (
.agents/blog-context.md) which may contain site metadata. If the agent is subsequently instructed to fetch an external URL (like a sitemap), there is a potential path for data leakage if not properly scoped.
Audit Metadata