elixir-vibe/pi-elixir/elixir
Claude Code and agent skill by elixir-vibe/pi-elixir
- community
Checking that the SKILL.md file is still in the repository.
Existing Elixir application/library development and debugging. Use for .ex/.exs, Mix, OTP, Ecto, backend Phoenix, dependencies, docs, tests, or BEAM runtime state. Use elixir-web instead when the primary task is UI/LiveView/frontend work, and elixir-new-project instead for creation/bootstrap.
Install
npx vskill@latest install elixir-vibe/pi-elixir/elixirSKILL.md
Elixir Development
Use this as the primary skill for existing general/backend Elixir work. Do not also load elixir-web unless the task genuinely spans substantial backend and UI work.
Default routing
- Runtime values, project APIs, dependencies, docs, processes, or typed data →
elixir_eval. - Elixir code shape, search, or refactoring →
elixir_ast_search/elixir_ast_replace. - Definitions, references, diagnostics, symbols, or code actions → LSP.
- Exact source reading or small textual edits → host
read/edit. - Git, external CLIs, and final Mix gates → shell.
Do not shell out to run Elixir snippets, grep for syntax when an AST pattern fits, or web-search before checking docs loaded in the BEAM.
Eval targets: project is the dependencyless default without application startup; application starts the managed app; runtime attaches to PI_ELIXIR_NODE; bridge hosts helpers such as AST, CodeMap, Self, Q, and Docs.
Work in a narrow loop: inspect, make the smallest appropriate edit, verify narrowly, then run project format/test/CI gates. Use AST.diff(changed: true) before large textual diffs and CodeMap.reflect(changed: true) when Reach is available.
Lazy references
Do not preload all references. Read only when the trigger applies:
references/tool-routing.md— tool choice is unclear, or the task involves a broad structural search/refactor.references/runtime-recipes.md— a concrete docs, OTP, Phoenix, Ecto, QuackDB, reload, or profiling recipe is needed.references/package-release.md— before publishing/releasing a Hex package or creating package release artifacts.
Before adding test support, inspect and extend the project's existing cases, drivers, assertions, fakes, fixtures, and naming/path conventions rather than creating a parallel hierarchy.
Security
- Security scanPassed, 95 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingpackages/extension/skills/elixir/SKILL.md
Popularity
- Repository stars
- 111
- Repository forks
- 7
- 7-day trend score
- 24
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/elixir-vibe/pi-elixir/elixir)