agent-web-compatibility
Pass
Audited by Gen Agent Trust Hub on Mar 26, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is composed of markdown-based instructions and reference files (SKILL.md, README.md, and the references/ directory). It contains no executable scripts or binaries.
- [SAFE]: No sensitive file access, credential harvesting, or unauthorized network operations were found in the instructions or templates.
- [PROMPT_INJECTION]: Analysis of the skill reveals a surface for indirect prompt injection, as it is designed to ingest and audit external website content.
- Ingestion points: Processes website URLs or codebase snippets provided by the user.
- Boundary markers: The skill employs a strict 3-layer framework (Discoverability, Preferability, Completability) which acts as a structured guide for the agent's reasoning.
- Capability inventory: The skill generates textual reports, JSON-LD markup, and llms.txt drafts. It does not have access to tools for file modification or network communication.
- Sanitization: None specified for input data, which is typical for this type of reporting tool. The risk is considered negligible as there are no exploitable capabilities.
Audit Metadata