frontend-ai-guide
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists exclusively of markdown documentation and YAML metadata. No scripts, binaries, or active code components are present in the skill package.
- [SAFE]: No network activity, external downloads, or remote code execution patterns were detected. The skill guides the agent to use existing repository-defined tools.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a standard operational surface for a code-review tool by instructing the agent to ingest and process untrusted frontend source code. 1. Ingestion points: The agent reviews React components, hooks, and browser behavior implementation based on the instructions. 2. Boundary markers: No explicit boundary markers or ignore-instructions are defined in the guide. 3. Capability inventory: The skill mentions executing local project-configured linting, type checking, building, and behavioral testing scripts (SKILL.md). 4. Sanitization: No sanitization or escaping of ingested code is specified for the prompt interpolation phase.
Audit Metadata