code-setup-project

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill performs command execution to manage the environment setup, including chmod +x to enable execution of wrapper scripts and nix flake check to validate the integrity of the generated Nix configuration.
  • [EXTERNAL_DOWNLOADS]: The provided Nix flake template includes dependencies on external sources, specifically the official NixOS/nixpkgs repository and the flake-utils library from numtide, both of which are hosted on GitHub.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted data from project configuration files to automate setup recommendations.
  • Ingestion points: Files like package.json, pyproject.toml, setup.py, go.mod, Cargo.toml, Gemfile, Makefile, and CMakeLists.txt are parsed for project type and dependency detection.
  • Boundary markers: The analysis process does not implement specific delimiters or instructions to ignore instructions embedded within the metadata of these files.
  • Capability inventory: The agent is capable of creating executable bash scripts, modifying file system permissions, and running Nix evaluation tools.
  • Sanitization: The skill lacks explicit sanitization or validation steps for the command strings and package names extracted from the ingested project files before using them in generated outputs.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 01:18 PM