nix-init
Pass
Audited by Gen Agent Trust Hub on Aug 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTIONREMOTE_CODE_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes shell commands using
nix runandnix-buildto generate, build, and verify Nix derivations. - [EXTERNAL_DOWNLOADS]: The skill uses
nix-initto fetch source code, metadata, and hashes from remote Git hosting services (e.g., GitHub, GitLab) to generate package definitions. - [PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection by ingesting and processing untrusted data from remote repositories to generate executable Nix code.
- Ingestion points: Remote repository URLs (SKILL.md).
- Boundary markers: None identified for delimiting untrusted content during the Nix build process.
- Capability inventory: Shell command execution via
nix-buildandnix run(SKILL.md). - Sanitization: None identified; the tool relies on inferred metadata from the remote source.
- [REMOTE_CODE_EXECUTION]: The skill generates Nix build scripts based on remote repository content and subsequently instructs the agent to execute these scripts using
nix-build, which constitutes dynamic execution of generated code.
Audit Metadata