Geepers
Multi-agent orchestration system with specialized Claude Code plugin agents and skills.
Installation
As Claude Code Plugin (recommended)
In Claude Code CLI:
# 1. Add the marketplace (one-time)
/plugin marketplace add lukeslp/geepers
# 2. Install the plugin
/plugin install geepers@geepers-marketplace
Manual Installation (skills only)
git clone https://github.com/lukeslp/geepers.git ~/geepers
mkdir -p ~/.claude/skills
for zip in ~/geepers/skills/zips/*.zip; do
name=$(basename "$zip" .zip)
mkdir -p ~/.claude/skills/$name
unzip -o "$zip" -d ~/.claude/skills/$name
done
What's Included
70 Specialized Agents
Markdown-defined agents for Claude Code that provide specialized workflows:
| Category | Agents | Purpose |
|---|---|---|
| Master | geepers_conductor | Intelligent routing to specialists |
| Checkpoint | scout, repo, status, snippets, orchestrator | Session maintenance |
| Deploy | caddy, services, validator, orchestrator | Infrastructure |
| Quality | a11y, perf, api, deps, critic, security, testing, orchestrator | Code audits |
| Frontend | css, design, motion, typescript, uxpert, webperf, orchestrator | UI/UX development |
| Fullstack | db, react, orchestrator | End-to-end features |
| Hive | builder, planner, integrator, quickwin, refactor, orchestrator | Task execution |
| Research | data, links, diag, citations, fetcher, searcher, orchestrator | Data gathering |
| Datavis | color, data, math, story, viz, orchestrator | Data visualization |
| Games | game, gamedev, godot, orchestrator | Game development |
| Corpus | corpus, corpus_ux, orchestrator | Linguistics/NLP |
| Web | flask, express, orchestrator | Web applications |
| Python | pycli, orchestrator | Python projects |
| Standalone | api, scalpel, dashboard, canary, janitor, docs, git, todoist, humanizer | Specialized tasks |
| System | help, onboard, diag | System utilities |
Packaged Skills
All agents are packaged as Claude Code skills in skills/zips/ (39 zips):
| Skill | Type | Description |
|---|---|---|
| conductor | Master | Top-level orchestrator routing to all agents |
| checkpoint | Orchestrator | Session boundaries, routine maintenance |
| deploy | Orchestrator | Deployment, infrastructure, Caddy |
| quality | Orchestrator | Code audits, a11y, perf, security, testing |
| frontend | Orchestrator | CSS, TypeScript, motion, design, UX |
| fullstack | Orchestrator | Backend + frontend end-to-end |
| hive | Orchestrator | Build from plans, execute backlogs |
| research | Orchestrator | Information gathering, API data |
| datavis | Orchestrator | D3.js, Chart.js, data visualization |
| datavis-agents | Orchestrator | Individual datavis specialist agents |
| games | Orchestrator | Game dev, Godot, gamification |
| corpus | Orchestrator | Linguistics, NLP, corpus work |
| web-dev | Orchestrator | Flask + Express web applications |
| python-dev | Orchestrator | Python projects, CLI tools |
| server-deploy | Standalone | dr.eamer.dev service management |
| engineering | Standalone | System architecture, full-stack code |
| executive | Standalone | Strategic planning, cross-team coordination |
| finance | Standalone | Monetization, financial modeling |
| product | Standalone | Product strategy, PRDs |
| api-design | Standalone | REST API design review |
| canary | Standalone | Quick health checks |
| dashboard | Standalone | System monitoring dashboard |
| docs | Standalone | Documentation generation |
| git-ops | Standalone | Git operations, conflict resolution |
| git-hygiene-guardian | Standalone | Git hygiene enforcement |
| janitor | Standalone | Project cleanup |
| scalpel | Standalone | Surgical code edits |
| todoist | Standalone | Todoist task management |
| humanize | Standalone | Remove AI writing indicators from documentation |
| system-diag | Standalone | Full system diagnostics |
| system-help | Standalone | Agent discovery and help |
| system-onboard | Standalone | Project onboarding |
| swarm | Research | Parallel multi-domain search |
| geepers-cascade | Research | Hierarchical deep research |
| geepers-publish | Workflow | Publishing and release workflow |
| geepers-team | Workflow | Team coordination workflow |
| data-fetch | MCP | Universal data fetching server |
| mcp-orchestration | MCP | Multi-agent orchestration server |
| vision | MCP | Alt text and image analysis |
Skill Structure
skills/
├── source/ # Skill source directories
│ ├── conductor/
│ │ └── SKILL.md
│ ├── quality/
│ │ ├── SKILL.md
│ │ └── agents/ # Sub-agent definitions
│ │ ├── geepers_a11y.md
│ │ ├── geepers_critic.md
│ │ └── ...
│ └── ...
├── zips/ # Packaged skill zips
├── rebuild-zips.sh # Rebuild all zips from source
└── package_all_skills.py # Generate skills from agents
Usage
Agents are available via Claude Code's Task tool with subagent_type:
# Quick reconnaissance
Task with subagent_type="geepers_scout"
# Infrastructure changes
Task with subagent_type="geepers_caddy"
# End-of-session cleanup
Task with subagent_type="geepers_orchestrator_checkpoint"
# Full-stack development
Task with subagent_type="geepers_orchestrator_fullstack"
# Frontend work
Task with subagent_type="geepers_orchestrator_frontend"
Agent Categories
See AGENT_DOMAINS.md for the full routing guide.
Orchestrators coordinate multiple specialists:
geepers_conductor- Master routergeepers_orchestrator_frontend- CSS, React, design, motiongeepers_orchestrator_fullstack- Backend + frontendgeepers_orchestrator_hive- Build from plans and TODOsgeepers_orchestrator_quality- Audits and reviewsgeepers_orchestrator_deploy- Infrastructure changesgeepers_orchestrator_checkpoint- Session maintenancegeepers_orchestrator_research- Data gatheringgeepers_orchestrator_datavis- Visualizationgeepers_orchestrator_games- Game developmentgeepers_orchestrator_corpus- Linguistics/NLPgeepers_orchestrator_web- Web applicationsgeepers_orchestrator_python- Python projects
Related: MCP-Dreamwalker
For programmatic orchestration with LLM providers and data clients, see MCP-Dreamwalker:
- Dream Cascade - Hierarchical 3-tier research orchestration
- Dream Swarm - Multi-domain parallel search
- 12 LLM Providers - Anthropic, OpenAI, xAI, Gemini, Mistral, Cohere, etc.
- 17 Data Clients - Census, arXiv, GitHub, NASA, Wikipedia, etc.
Building Skills
# Regenerate all SKILL.md files from agent definitions
python3 skills/package_all_skills.py
# Rebuild all zip files
bash skills/rebuild-zips.sh
# Install to Claude Code
for zip in skills/zips/*.zip; do
name=$(basename "$zip" .zip)
mkdir -p ~/.claude/skills/$name
unzip -o "$zip" -d ~/.claude/skills/$name
done
License
MIT License - Luke Steuber
Metadata
Release files for conductor-geepers 1.0.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| conductor_geepers-1.0.1.tar.gz | 273.5 kB | Without provenance | Details for conductor_geepers-1.0.1.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| conductor_geepers-1.0.1-py3-none-any.whl | Python 3 | none | any | Details for conductor_geepers-1.0.1-py3-none-any.whl |
Total release size: 418.4 kB
Release files / conductor_geepers-1.0.1.tar.gz
| Download URL | conductor_geepers-1.0.1.tar.gz |
|---|---|
| Size | 273.5 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
877bc73bbf61daec444e6108fc4b7f1de13d8a97a33d81331062a93413322177
|
|
BLAKE2b-256 checksum How to use checksums |
94ef19c88af7aedee1a9ed79afb7b2862bb19cf46b41b04ccac406ea8b0efabf
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.12.3
|
Release files / conductor_geepers-1.0.1-py3-none-any.whl
| Download URL | conductor_geepers-1.0.1-py3-none-any.whl |
|---|---|
| Size | 144.9 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
17421f7def68e792cb262e1a789d4b399ad77754241f4d4e215fb73b1957c000
|
|
BLAKE2b-256 checksum How to use checksums |
7ccbd6cfaf318a2030371760dce38744601602ca05566cb217d4bdfa78a102f7
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.12.3
|
Security for conductor-geepers 1.0.1
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.