Back to directory

OneCLI

An open-source platform that gives every employee their own sandboxed AI agent, provisioned from the company identity provider and governed by one team-wide policy.

OneCLI

OneCLI is an open-source platform for running AI agents as a team. It creates one agent per person, gives each agent only the access it needs, and runs it in a sandbox reached through a gateway that injects credentials and enforces the organization's policy.

The project began as a Rust-based credential vault for autonomous agents such as Hermes and OpenClaw, then expanded to add team features: provisioning agents from a company identity provider, a durable per-agent sandbox with its own filesystem and shell reachable only through the gateway, per-agent conversations in a dashboard or Slack, memory, deterministic human-in-the-loop approval for sensitive actions such as sending an email or emptying a storage bucket, and shared team-level connections such as LLM keys granted per agent without ever being handed to it directly.

It can be run self-hosted, with pnpm, or used as a cloud-hosted service.

More in Agent frameworks

See all 902
  • Recurse

    recurse.run

    A serverless harness that lets a coding agent design, test, and deploy its own specialist sub-agents as tools, MCP servers, or bots, defined by a schema-validated manifest.

  • ProofForge

    shulgin.is-a.dev

    An AI-agent pipeline that decomposes a mathematics problem, proves each piece, and formalizes the result in Lean 4 and Mathlib until the Lean kernel accepts it, with six of its proofs already merged into Google DeepMind's formal-conjectures repository.

  • local-coder

    github.com

    Builds a local coding-agent team on top of Ollama and OpenCode by detecting the machine's hardware, recommending models for seven specialist roles, and testing whether each assigned model can actually use tools, edit a repository and pass verification.