improve-content
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from external URLs to perform analysis and rewriting, which creates a surface for indirect prompt injection where malicious instructions embedded in a target web page could influence the agent's behavior.
- Ingestion points: Fetches full rendered pages from user-supplied URLs in Step 1 and accepts manually pasted content.
- Boundary markers: The instructions do not define clear delimiters or provide specific warnings to the agent to disregard instructions or adversarial text found within the fetched content.
- Capability inventory: The agent has the ability to fetch web content, perform Google searches, and generate Markdown output.
- Sanitization: No specific sanitization or validation logic is specified for the external data ingested by the agent.
Audit Metadata