third-party-integrations
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to fetch and process external content from the Val Town documentation sitemap (
https://docs.val.town/sitemap-0.xml) and specific guide pages. This creates a potential surface where malicious instructions could be embedded in the fetched documentation. - Ingestion points: Markdown guide pages and XML sitemap from
docs.val.town(identified in SKILL.md). - Boundary markers: The skill does not define specific delimiters for separating documentation content from instruction context.
- Capability inventory: The agent generates integration code based on the fetched content, which may include network requests, database operations, or browser automation.
- Sanitization: No explicit instructions for sanitizing or escaping the documentation content before processing are included.
- [SAFE]: The skill promotes secure credential management by instructing the agent to use environment variables (
Deno.env.get) and providing direct links to the platform's native secret management interface.
Audit Metadata