Plugin Development Guide
Development documentation for the Elixir/Phoenix Claude Code plugin.
## Overview
This plugin provides **agentic workflow orchestration** with specialist agents and reference skills for Elixir/Phoenix/LiveView development.
## Workflow Architecture
entry without a `git tag` is the **current tag**, and the per-tag root documents (`PLAN.md`, `STRUCTURE.md`, `SCENES.md`) are scoped to it. `gm-finalize` archives the tag's working docs
Code (claude.ai/code) when working with code in this repository.
## Overview
This is a **documentation and template repository** for Claude Code best practices. It contains:
- Educational guide (GUIDE.md) explaining Claude
Commit with list of added repositories
### Reorganizing Categories
If Daniel requests category restructuring:
1. Document the new category structure
2. Create new category files or rename existing ones (maintain numeric
single-source rule lives in [`docs/changelog-fragments.md`](docs/changelog-fragments.md)
and is summarized in [`CONTRIBUTING.md`](CONTRIBUTING.md) (the "Documentation gate" section
onto_ossie_import` | To compile an Apache Ossie (incubating, formerly Open Semantic Interchange) ontology document into OWL 2 DL + SHACL and optionally load it, making a vendor semantic model reasonable/validatable
skill changes, update the docs in the same commit.
`docs/prior-art.md` is the exception — it documents adjacent ideas and trade-offs. Update when adding a scenario that explicitly contrasts with
signal synthesis is better than exhaustive capture. This system is for thinking, not for documenting everything.
## Routing
Start every task at [`INDEX.md`](./INDEX.md). It routes to every area. Strategy lives
every secret-scanning alert individually.** For an OAuth installed-app
identifier, verify from provider documentation whether it is intentionally public
and whether reuse is permitted. Record the source, scope, authorization
complementary state management patterns:
#### 1. URL-based State Management
See `@docs/url-state-management.md` for detailed documentation.
- Navigation state (tabs, views) is stored in URL parameters
- Enables shareable URLs, browser history, and bookmarking
supplementary docs loaded on demand
*.md
scripts/ # Optional: R or shell helpers
templates/ # Optional: document templates
```
Do **not** create a `README.md` inside individual skill directories — documentation about a skill
check if a lesson exists before fixing
2. **Read the lesson** — apply only the documented fix
3. **Submit gaps** — if no lesson found, submit via MCP intake
4. **Never skip
Standards
- No `console.log` in production code
- Always handle loading and error states
- Write self-documenting code
- Keep functions focused and small
- Use TypeScript strictly
## Key Development Principles
1. **Reuse First
when working with code in this repository.
## Repository Information
- Repository URL: https://github.com/1mcp-app/agent
- Documentation: https://docs.1mcp.app
## Development Commands
```bash
# Setup
pnpm install
cp .env.example .env # Required for development
# Build
cargo check`, if you have spare hardware reachable over SSH.
Specific hostnames/users/specs are NOT documented here — this is a public
repo, and machine names + SSH access patterns are reconnaissance info
A file Claude Code reads at the start of every session. It holds the commands, conventions and warnings the agent needs for this project.
Where does it go?
At the repository root. Claude Code also reads CLAUDE.md files in subdirectories when it works there.
What should it contain?
Build and test commands, the project's layout, conventions that aren't obvious from the code, and mistakes to avoid. Short files tend to work better than long ones.
CLAUDE.md or AGENTS.md?
Claude Code reads CLAUDE.md; most other agents read AGENTS.md. Many projects keep one and point the other at it.