neovim-plugin-development

Pass

Audited by Gen Agent Trust Hub on Sep 28, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to configure GitHub Actions workflows using community-standard resources such as actions/checkout, NTBBloodbath/selene-action, JohnnyMorganz/stylua-action, rhysd/action-setup-vim, and softprops/action-gh-release. It also pulls nvim-lua/plenary.nvim as a test dependency.
  • [COMMAND_EXECUTION]: The skill utilizes Makefile targets to execute Neovim in headless mode (nvim --headless) for running Lua tests and generating help documentation, which is the conventional method for Neovim plugin automation.
  • [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection via user-supplied requirements.
  • Ingestion points: User-provided plugin names, feature requests, and configuration options during scaffolding and editing tasks.
  • Boundary markers: Absent. The skill does not prescribe the use of delimiters or ignore-instructions for user data.
  • Capability inventory: The skill generates file structures (Lua modules, workflows) and shell-executable Makefiles.
  • Sanitization: Absent. No explicit sanitization or validation of user-provided content is required before generating code.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 28, 2026, 05:29 PM