ESLint Rule Generator Agent
Pass
Audited by Gen Agent Trust Hub on Mar 28, 2026
Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill consists of a single documentation file (SKILL.md) with no accompanying executable scripts or binaries. All behavior described is purely informational.\n- [PROMPT_INJECTION]: The skill describes a capability to translate untrusted natural language descriptions into executable code (ESLint rules). Ingestion points: Natural language rule specifications provided by users. Boundary markers: None specified in the documentation. Capability inventory: Generates code (ESLint rules) and supports publishing as npm plugins. Sanitization: None mentioned. This identifies a surface area for indirect prompt injection, although no malicious intent is evident in the documentation.\n- [EXTERNAL_DOWNLOADS]: The skill references source code from the official ESLint GitHub repository and installation via standard package managers. These references target well-known and author-owned repositories consistent with developer workflows.\n- [COMMAND_EXECUTION]: No unauthorized or suspicious command execution patterns are present. The documented functionality involves standard development tasks like AST parsing and rule testing within the ESLint ecosystem.
Audit Metadata