By cutalion
Universal multi-agent engineering team — analyst, planner, plan-critic, engineer, code-critic, QA, and writer working in sequence. Run /fixiki:team <task> to ship a feature branch from a fuzzy ask.
Use when an incoming task is vague, ambiguous, or lacks acceptance criteria. Turns a fuzzy ask into a clear mini-spec written to .ai_team/specs/<slug>.md. Owns "what & why." May ask the user clarifying questions in sync mode; in async mode, writes open questions into the spec and stops. Does NOT write plans, code, or implementation details.
Use after team-analyst produces a spec for a task that introduces or changes a public API, data model, cross-component contract, or non-trivial architecture. Produces a design doc at the charter's design path, drafts ADRs at the charter's ADR path (writer finalizes after merge), and drafts API contracts at the charter's contracts path if declared. Owns "what shape." Does NOT write file-by-file plans (that's planner) or code (that's engineer).
Use after team-engineer claims a task is done, before team-qa. Reviews uncommitted or recently-committed changes for bugs, logic flaws, style violations, and inconsistency with surrounding code. Read-only. Returns approve / request-changes with line-cited findings.
Use on /fixiki:team curate, before team-writer finalizes a PR, and in summary-only mode for /fixiki:team status. Audits docs at the charter's declared paths for staleness, supersession, and orphaning. Writes status frontmatter changes only — does not rewrite content. Returns a drift report; the lead routes content fixes to writer/architect.
Use when a task introduces or changes business rules, domain concepts, or ubiquitous-language terms — AND the charter declares a `Domain rules` path. Maintains the project's domain doc, captures invariants, names concepts. Owns "what the business means." Does NOT write architecture (that's architect), plans (planner), or code (engineer).
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
A universal multi-agent engineering team for Claude Code. Drop it into any repo and invoke /team <task> — seven specialist agents coordinate to ship a feature branch from a fuzzy ask.
Named after the Fixies — small creatures who fix things.
/fixiki:team init Scaffold .ai_team/ in your repo (one-time setup)
/fixiki:team <task> Run the team on a free-form task
/fixiki:team status Print current state and recent log
The current Claude Code session becomes the lead. It reads your charter, decides which agents to dispatch, and synthesizes one coherent report back to you.
| Agent | Role |
|---|---|
team-analyst | Turns a fuzzy ask into a clear spec with acceptance criteria |
team-planner | Turns a spec into a numbered, TDD-shaped implementation plan |
team-plan-critic | Reviews the plan before any code is written |
team-engineer | Implements on a feature branch using TDD |
team-code-critic | Reviews the implementation for bugs, style, and consistency |
team-qa | Runs project-appropriate test + lint gates |
team-writer | Updates docs and drafts PR descriptions |
The lead orchestrates them in sequence, runs review loops (up to 2 rounds each), and escalates to you when stuck.
In a Claude Code session, add the marketplace and install the plugin:
/plugin marketplace add cutalion/fixiki
/plugin install fixiki@fixiki
Then scaffold the team state in your project:
/fixiki:team init
This creates .ai_team/charter.md with auto-detected project name, language, test command, and lint command. Edit the charter to set your goals and authority level before running the team on a task.
.ai_team/charter.md is read on every invocation. Key settings:
## Goals
1. Ship feature X by date Y
## Authority
- Default scope: branch-only # engineer works on feature branches only
# Change to `broad` to allow auto-merging low-risk classes
## Project conventions
- Language: JavaScript (Node.js) # auto-detected; override if wrong
- Test command: npm test
- Lint command: npm run lint
/fixiki:team add dark mode to the settings page..ai_team/charter.md and the project structure.request-changes triggers one re-dispatch of the upstream agent (2-round limit)..ai_team/ travels with the repo:
.ai_team/
├── charter.md — you edit this
├── state.yml — the lead edits this
├── log/ — one file per session
└── specs/ — analyst-authored mini-specs
/team LIN-123 pulls issue body automatically./team start runs the team on a schedule via /schedule.MIT.
npx claudepluginhub cutalion/fixiki --plugin fixikiUnity Development Toolkit - Expert agents for scripting/refactoring/optimization, script templates, and Agent Skills for Unity C# development
Complete creative writing suite with 10 specialized agents covering the full writing process: research gathering, character development, story architecture, world-building, dialogue coaching, editing/review, outlining, content strategy, believability auditing, and prose style/voice analysis. Includes genre-specific guides, templates, and quality checklists.
Comprehensive .NET development skills for modern C#, ASP.NET, MAUI, Blazor, Aspire, EF Core, Native AOT, testing, security, performance optimization, CI/CD, and cloud-native applications
Complete collection of battle-tested Claude Code configs from an Anthropic hackathon winner - agents, skills, hooks, and rules evolved over 10+ months of intensive daily use
Modern R development skills for Claude Code - tidyverse patterns, rlang metaprogramming, Bayesian inference, performance optimization, and more
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.