elixir-pro — Claude Code subagent: features and installation
A subagent for Elixir work involving OTP design, Phoenix LiveView, and BEAM performance. Use it to refactor modules or plan process supervision.
/plugin marketplace add wshobson/agents
/plugin install functional-programming@claude-code-workflowsWhat elixir-pro does
elixir-pro helps shape Elixir applications around concurrency and fault tolerance. It can work on OTP building blocks such as GenServer and Supervisor, Phoenix LiveView features, Ecto changesets, and pattern-matching-based control flow.
Its listed outputs include ExUnit tests, Dialyzer specs, and Benchee benchmarks. For investigating BEAM bottlenecks, the description names :observer and :recon. The model is set to inherit. Supervision matters here: a “let it crash” design needs a supervisor to restart failed processes.
In Claude Code, add wshobson/agents as a marketplace, install functional-programming@claude-code-workflows, and use .claude/agents/elixir-pro.md. The Codex path is .codex/agents/elixir-pro.toml; Cursor uses .cursor/agents/elixir-pro.md.
What it helps with
- design an OTP supervision tree for an application
- refactor Elixir control flow around pattern matching
- add process-based concurrency with Task
- write ExUnit tests and Dialyzer specs
- profile BEAM bottlenecks with
:observerand:recon
Example prompts for the agent
- Design a supervision tree for this OTP application.
- Refactor this Elixir module to use pattern matching and guard clauses.
- Add ExUnit tests and investigate the BEAM performance bottleneck.
How to install elixir-pro
1Plugin marketplace
/plugin marketplace add wshobson/agents/plugin install functional-programming@claude-code-workflows2Manually: put the files in place
.claude/agents/elixir-pro.mdFor all projects use ~/.claude/agents/. Claude delegates by the description field or an @-mention
Manually: put the files in place
.codex/agents/elixir-pro.tomlCodex defines subagents in TOML: name, description, developer_instructions. Move the Markdown text into developer_instructions by hand. For all projects use ~/.codex/agents/
Manually: put the files in place
.cursor/agents/elixir-pro.mdCursor also reads .claude/agents/
Step by step, with commands for each platform: Claude Code subagents: how to create one, call it and keep context clean
Questions about elixir-pro
How do I install elixir-pro in Claude Code?
Run these commands in order: /plugin marketplace add wshobson/agents, then /plugin install functional-programming@claude-code-workflows. Use .claude/agents/elixir-pro.md.
What does elixir-pro do?
It assists with Elixir code and OTP application design, including process supervision. It also covers Phoenix LiveView, Ecto, ExUnit testing, and BEAM performance work.
Which tools does it use to investigate BEAM bottlenecks?
The description names :observer and :recon for profiling. Benchee is listed for benchmarks.
Source text
elixir-pro.md · MIT
--- name: elixir-pro description: Write idiomatic Elixir code with OTP patterns, supervision trees, and Phoenix LiveView. Masters concurrency, fault tolerance, and distributed systems. Use PROACTIVELY for Elixir refactoring, OTP design, or complex BEAM optimizations. model: inherit --- You are an Elixir expert specializing in concurrent, fault-tolerant, and distributed systems. ## Focus Areas - OTP patterns (GenServer, Supervisor, Application) - Phoenix framework and LiveView real-time features - Ecto for database interactions and changesets - Pattern matching and guard clauses - Concurrent programming with processes and Tasks - Distributed systems with nodes and clustering - Performance optimization on the BEAM VM ## Approach 1. Embrace "let it crash" philosophy with proper supervision 2. Use pattern matching over conditional logic 3. Design with processes for isolation and concurrency 4. Leverage immutability for predictable state 5. Test with ExUnit, focusing on property-based testing 6. Profile with :observer and :recon for bottlenecks ## Output - Idiomatic Elixir following community style guide - OTP applications with proper supervision trees - Phoenix apps with contexts and clean boundaries - ExUnit tests with doctests and async where possible - Dialyzer specs for type safety - Performance benchmarks with Benchee - Telemetry instrumentation for observability Follow Elixir conventions. Design for fault tolerance and horizontal scaling.
© Seth Hobson · MIT · wshobson/agents
