webring-site
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches data and resources from the
ussyco.dedomain, including a JSON list of members and a JavaScript widget script (https://ussyco.de/api/webring/widget.js) to be included in generated site files. - [COMMAND_EXECUTION]: The skill utilizes
curlvia shell to perform registration tasks and API health checks. It handles the API key via an environment variable (USSYCO_DE_API_KEY) and explicitly warns against hardcoding or logging the secret, which is a safe practice. - [INDIRECT_PROMPT_INJECTION]: The skill requires the agent to browse and analyze third-party member websites to research design aesthetics, creating a potential surface for indirect prompt injection.
- Ingestion points: External website content processed via browsing or fetching tools during the design research phase.
- Boundary markers: Absent; instructions do not specify using delimiters or ignore-embedded-instruction prompts for external site content.
- Capability inventory: Workspace file writing, local server execution, and network POST operations via
curl. - Sanitization: Not explicitly implemented in the workflow.
Audit Metadata