Fenix Agent Plugin
Structured workflows and specialized agents for AI coding tools
Your AI agent has 68+ tools but no idea when to use them. Fenix adds the methodology layer — structured workflows, specialized agents, cross-session memory, and team-specific skills — so your agent plans before coding, tests before shipping, and remembers what happened yesterday.
Authentication is handled via OAuth 2.1 — your browser opens once for login, and tokens are managed automatically. No API keys or PATs to configure.
What you get
| Feature | Description |
|---|
| Agents | Specialized agents | Fenix Plan, Build, Review, Deliver — each focused on one workflow phase |
| Planning | Discussion phase | Gray areas identified, options presented with trade-offs, decisions captured |
| Memory | Cross-session continuity | Agent searches past context at start, saves session state at end |
| TDD | Test-driven development | Red-Green-Refactor with automatic detection of when TDD applies |
| Skills | Dynamic discovery | Agents find and follow your team's standards automatically — no manual config |
| Standards | Guided creation | No team standards? Agent suggests creating them with your chosen scope |
| Boards | Status progression | Correct transitions: backlog → todo → in_progress → review → done |
| Context | Auto-loaded profile | User context, team info, and preferences injected before first message |
| Language | Multi-language | Agent responds in your language — configs are English, conversations are yours |
Installation
Claude Code
Step 1 — Install the plugin:
/plugin marketplace add fenix-assistant/fenix-plugin
/plugin install fenix
/reload-plugins
Step 2 — Authenticate:
- Run
/plugin to open the plugin panel
- Find Fenix MCP Server and click Connect
- Your browser opens for OAuth login — sign in with your Fenix account, select a team, and approve
- Done! The MCP server connects automatically
No PAT needed. Authentication is handled via OAuth 2.1 — your browser opens once, and tokens are cached automatically.
After installing, you get:
- OAuth authentication — browser-based login, no tokens to manage
- Bootstrap hook — injects workflow methodology at session start
- 4 subagents — planner, builder, reviewer, deliverer (delegated via Agent tool)
- 11 skills — loaded on-demand by the orchestrator
Multiple Fenix accounts (multi-tenant)
If you use different Fenix accounts in different projects (e.g., personal account in ~/dev/myproject and work account in ~/dev/company), install the plugin with the "local" scope (per-project). Each project directory automatically gets its own OAuth session — no extra configuration needed.
- Open Claude Code in each project directory
- Install with local scope:
/plugin install fenix (select "local" when prompted)
- Run
/reload-plugins
- Go to
/plugin → Connect on the Fenix MCP Server
- Log in with the Fenix account you want for that project
The plugin automatically generates a unique identifier per project directory, so each project authenticates independently. You can use a different Fenix account in each project without conflicts.
Single account users don't need this. If you only use one Fenix account, just install and authenticate — it works out of the box.
To update: /plugin update fenix then /reload-plugins
OpenCode
Paste this into your agent:
Install fenix-plugin following: https://raw.githubusercontent.com/fenix-assistant/fenix-plugin/main/docs/install-opencode.md
The agent will ask for your PAT, configure the MCP server, and install the fenix-opencode plugin.
fenix-assistant