direnv

Pass

Audited by Gen Agent Trust Hub on Jul 8, 2026

Risk Level: SAFE
Full Analysis
  • [PRIVILEGE_ESCALATION]: Documentation in references/install-and-setup.md includes standard installation instructions for Linux systems using sudo with system package managers such as apt and dnf.
  • [PERSISTENCE_MECHANISMS]: The skill provides instructions for adding shell hooks to user profile files (e.g., ~/.bashrc, ~/.zshrc) to enable automatic environment variable loading, which is the standard setup for the tool.
  • [INDIRECT_PROMPT_INJECTION]:
  • Ingestion points: The skill documents the processing of .envrc files located in project directories (SKILL.md).
  • Boundary markers: Includes a 'Safety Guardrails' section explicitly warning users to review file contents before running direnv allow to prevent execution of untrusted code (SKILL.md).
  • Capability inventory: The tool manages environment variable exports and can execute scripts via the .envrc file and the direnv exec command (references/command-cookbook.md).
  • Sanitization: Implements a security model that blocks execution of new or modified configuration files until the user manually approves them using direnv allow (references/install-and-setup.md).
  • [DYNAMIC_EXECUTION]: Uses eval to initialize shell hooks within profile files, which is a standard method for integrating the tool's functionality into the shell environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 8, 2026, 04:18 PM
Security Audit — agent-trust-hub — direnv