Stars
Forks
Watchers
Developer links
Opengeni
Enterprises requiring production-grade governance for long-running autonomous AI operations can rely on OpenGeni, a self-hosted orchestration platform with human-in-the-loop approvals, Connected Machine execution, and comprehensive audit replays. Operators can configure durable multi-turn workflows that execute continuously toward specified goals without premature termination, automatically pausing when human verification, structured multiple-choice decisions, or critical credential approvals are required. Teams can assign execution workloads across ephemeral cloud sandboxes or route them straight to enrolled Connected Machines to run commands directly against local code repositories without exposing inbound network ports. The integrated web console lets developers inspect live Server-Sent Event activity streams, replay complete turn histories from audit logs, inspect generated artifacts, and cancel or steer in-flight agent tasks on demand. Platform administrators can enforce fine-grained access policies through GitHub App repository bindings, broker temporary role-scoped secrets, and maintain a shared organizational knowledge base where agents propose findings for human review before vector indexing. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache-2.0 licensed.
Benefits
- Durable Replayable Agent Session State
- Stores every execution event in append-only PostgreSQL tables, enabling real-time Server-Sent Event streaming with automatic backfill so browser reloads and audits replay exact turn histories without data loss.
- First-Class Dual Compute Target Architecture
- Executes agent workloads inside ephemeral managed sandboxes or directly on enrolled Connected Machines that dial out via secure relays, eliminating inbound network exposure and cloud token transfers.
- Human-In-The-Loop Approval and Steering Controls
- Gates sensitive tool executions behind mandatory operator confirmations, supports structured multiple-choice questionnaire pauses, and allows real-time prompt steering to redirect running tasks without dropping existing progress.
- Governed Knowledge and Proposed Memory
- Indexes enterprise documentation using pgvector hybrid search and requires explicit human or API verification before agent-discovered facts and procedural memories become persistent retrieval context across sessions.
Features
- Temporal Orchestration Engine
- Coordinates multi-turn agent activities, retries, and signaling workflows using Temporal to guarantee fault-tolerant execution across infrastructure outages and worker restarts.
- Connected Machine Rust Relay
- Connects local development machines and build hosts to the control plane using an outbound-only Rust agent and NATS message bus without opening inbound firewall ports.
- Goal-Oriented Autonomous Execution
- Equips sessions with explicit success criteria and budget guards that prevent premature completion until verifiable evidence is submitted through explicit goal fulfillment tools.
- Vector Knowledge Search
- Processes uploaded documentation, repository code, and structured findings with PostgreSQL pgvector embeddings to provide high-precision semantic retrieval for reasoning workflows.
- Interactive React Console
- Delivers a Vite-based web interface featuring real-time Server-Sent Event timelines, terminal stream relays, artifact inspection viewers, and granular workspace management.
- Granular GitHub App Security
- Mints short-lived, installation-scoped tokens through an official GitHub App integration to enforce least-privilege repository checkouts and secure pull request automation.