Practical overview
When it fits
Use it when work must remain understandable and reusable after an individual agent session ends, especially across Claude Code and Codex sessions.
Who it is for
- Teams coordinating multiple coding-agent sessions
- Developers using Claude Code, Codex or both on shared projects
- Projects that need durable task history, memory, reports and handoffs
Documented uses
- Share tasks, ownership, blockers and completion records across sessions
- Preserve project decisions, preferences, reports and task memos
- Send and receive project-scoped messages and structured coordination signals
- Monitor agent work and project activity in a dashboard
- Connect native coding agents through MCP-based coordination
Requirements & limits
The repository is public, open source and licensed under MIT.
- Claude Code and Codex can connect to the same OS resources; this is documented as Verified.
- Execution remains native to each host, with separate permissions, settings, schedulers and lifecycle handling.
- Claude Code has documented hook-based briefing and observation extensions that are not Codex features.
- MCP tooling is evidenced; use with an additional MCP-compatible host is marked Likely compatible, not demonstrated.
- The project uses Python, MCP, FastAPI and React and provides a dashboard for recorded work and activity.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Connect Claude Code and Codex to shared project records
Allow both native agent hosts to use the same task wall, project memory, reports, channels and dashboard.
Before you start
- Claude Code and/or Codex configured for AI Team OS
- A shared project and the hosts’ respective MCP and hook configuration
Steps
Connect Claude Code and Codex to the same AI Team OS resources.
Use the shared task wall, project memory, reports and channels for coordination.
Record progress, decisions and reports through the shared MCP tools.
Inspect the Dashboard to compare recorded work with current activity.
Expected outcome
Both hosts can contribute to shared project records while retaining separate native tools, settings and lifecycle boundaries.
Guide limitations
- Codex uses its own MCP and hook configuration.
- Native agent tools, host settings and hook trust remain separate.
- The operating layer does not replace either host’s scheduler, permissions or agent lifecycle.
Compare options for Agent orchestration
These projects document a common task and share a project role. This is a source-based comparison, not an execution benchmark or a guarantee of interchangeability.
| Project | Documented purpose | Platform | Access and requirements |
|---|---|---|---|
| AI Team OS | Share tasks, ownership, blockers and completion records across sessions | CLI, Self-hosted, Server, Web | The repository is public, open source and licensed under MIT. |
| Massgen | Run multiple AI agents against the same user task. | CLI | The repository documents installation through pip or uv and requires Python 3.11 or later. Setup can configure API keys for OpenAI, Anthropic, Google and xAI; credentials are needed for the selected model backends. |
| Langchain Multi Agents Boilerplate | Route incoming football questions to model and agent workflows according to task complexity. | Web | The repository is public and specifies an MIT license. Running the application requires OpenAI and Groq API keys and a Redis server. |
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 30 September 2026
- Original website title
- CronusL-1141/AI-company · Official homepage title · retained in its source language