nix-language

Pass

Audited by Gen Agent Trust Hub on Oct 10, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze and evaluate external Nix expressions from a user's project, which represents a potential attack surface for indirect prompt injection if those expressions contain malicious instructions. The skill proactively mitigates this risk by providing explicit instructions to the agent to inspect code before execution.
  • Ingestion points: Nix expressions and project files read by the agent (instructed in SKILL.md).
  • Boundary markers: The skill directs the agent to "Inspect unknown expressions before executing them and respect the task's existing permissions."
  • Capability inventory: The Nix evaluator has capabilities to read files (builtins.readFile), access environment variables (builtins.getEnv), and fetch external resources (builtins.import, derivations).
  • Sanitization: The skill relies on the agent's manual inspection and adherence to project-specific permissions as the primary sanitization mechanism.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 10, 2026, 03:53 PM