The Agent Skills Directory

[SAFE]: The skill documents standard Nix ecosystem patterns and best practices with no evidence of malicious intent or dangerous capabilities.
[EXTERNAL_DOWNLOADS]: Documents legitimate Nix functions for fetching source code (fetchFromGitHub, fetchurl) and includes an example utility using curl to fetch gitignore templates from the well-known service gitignore.io. Importantly, the skill includes explicit warnings against using these tools without integrity hashes.
[COMMAND_EXECUTION]: Provides patterns for shell-based build phases in Nix derivations (installPhase) and shell hooks for development environments (shellHook). These are standard Nix features used according to established ecosystem conventions.
[PROMPT_INJECTION]: No attempts to override agent behavior, bypass safety guidelines, or extract system prompts were detected.
[DATA_EXFILTRATION]: No sensitive file access or unauthorized network operations were identified. Credential management examples (e.g., GPG signing keys in Git configuration) follow standard practice without hardcoding secrets.
[SAFE]: The skill explicitly promotes security-positive behaviors, such as avoiding impure absolute paths and ensuring reproducibility through pinning and lock files.

Nix Ecosystem