BigHugger
GH Repository · Devin-AXIS

iPolloWork

Enterprise-grade, local-first Agent Workbench for people and agent teams. A unified multi-engine workspace for Codex Harness, DeepSeek Harness, and OpenCode, with unified plugins and Skills, multi-agent projects and tasks, and editable code, documents, presentations, design, and video.

stars
6,080
30-day movement
+413138/day
Related entries
60
Connections
3
node/pnpmagent-appenterprise-agent-workspaceai-agentsmulti-agent-workspacecodexcodex-pluginopencodeipollowork-skilldeepseek-harnessTypeScriptself-evolving-aidsh-pluginvisual-editoropencode-pluginagent-skillsipollowork-pluginclaude-codenodehermes-agentai-workagent-collaboration

iPolloWork is a local-first Agent Workbench that provides a unified workspace for multiple agent engines, including Codex Harness, DeepSeek Harness, and OpenCode. It combines shared plugins and Skills, multi-agent project and task management, and editable code, documents, presentations, design, and video in one app. It is written in TypeScript and ships with agents.md and a devcontainer.

Reach for it when you want to run and coordinate several agent engines, their plugins, and the work they produce from a single local workspace instead of separate tools.

Use it to

  • Run Codex, DeepSeek, and OpenCode engines from one workspace
  • Manage multi-agent projects and shared tasks
  • Share plugins and Skills across agent engines
  • Edit agent-produced code, documents, and presentations
  • Set up a reproducible environment with the shipped devcontainer

For Developers and teams running multiple AI agent engines

Role
agent-app
Language
TypeScript
Licence
custom (licence file present)
Forks
1,161
Open issues
63
Last push
2026-09-15
Latest release
alpha-macos-latest · 2026-07-20
topicsagent-workbenchmulti-agentopencodecodexdeepseektypescript