Self-host Jentic One
v0.44.0Execution layer securing AI agent API calls
- Stars
- 216
- Port
- 8000 (control plane (UI + APIs))
ghcr.io/jentic/jentic-one-appWhat is Jentic One?
Jentic One is a self-hosted platform that sits between AI agents and external APIs. Instead of giving agents API keys directly, operators register APIs and store credentials once in Jentic One. Agents then make calls through the Jentic Broker, which checks permissions, injects the appropriate credential at execution time, and records every request. The system deploys as two peer services — an App (control plane with UI, registry, auth, and admin) and a Broker (data plane that injects credentials) — backed by PostgreSQL or SQLite. It integrates with coding agents like Claude Code, Codex, and Cursor via a local MCP server or CLI, and supports OIDC for single sign-on. The project is in public beta (0.x releases) with schemas and CLI commands subject to change.
Should you self-host Jentic One?
Jentic One scores 48/100 on selfhost.directory (how scoring works) and ranks #110 of 209 in AI Agents. It has 216 GitHub stars.
It shipped 5 stable releases in the past year. The latest, v0.44.0, came out on 27 October 2026. New versions land roughly every 1 days.
Setup is moderate: expect some configuration beyond a single command. An official Docker image is available. It is Apache-2.0-licensed and written mostly in Python.
Its closest alternative on selfhost.directory is Orca (91/100, 108 446 stars). It scores higher overall, so compare the two before you pick. See Jentic One vs Orca side by side.
Features
- Agent CLI
- Anonymous Telemetry (Opt-in)Enabled by default in quickstart config; can be set to false.
- API AI-Readiness Scoring
- API Catalog with 10,000+ APIs
- Arazzo Workflow Engine
- Credential Broker
- Credential Rotation & Revocation
- Default-Deny Policy Engine
- Execution Logging & Observability
- MCP Server
- Multi-database Support
- OIDC Authentication
- Operator CLI
- Require-Approval Workflows
Jentic One README
FAQ
Does Jentic store execution logs?
Yes, for observability. Every brokered call writes an execution record, visible under Monitor → Executions in your instance.
Can I rotate credentials?
Yes — credentials and agent tokens can be revoked and regenerated at any time from your Jentic One dashboard.
Is my data secure?
Yes. API credentials are stored encrypted in your Jentic One Broker and injected at execution time — never sent to the LLM or returned to a caller. Access is default-deny.
Do you support MCP?
Yes. Search the public API directory over the hosted MCP server, and connect your own Jentic One instance as a custom connector for the full search → inspect → execute loop in tools like Claude, Windsurf, Cursor, and ChatGPT.
How do agents call Jentic?
Through your self-hosted Jentic One instance via the jentic CLI (search → inspect → execute) or over MCP.
How can I contribute an API or workflow?
Post your OpenAPI spec link in the #api-requests Discord channel or open an issue at jentic/jentic-public-apis using the import template.
Do I paste raw API secrets into prompts?
Never. Credentials are stored encrypted in your Jentic One Broker and injected at execution time — they never reach the agent, its prompts, or the response.
Can I browse APIs without an account?
Yes, the full catalog is the open-source Jentic Public APIs knowledge base, searchable in natural language over the hosted MCP directory with no account required.
What is API Search?
API Search is the in-app catalog of real-world capabilities from thousands of APIs, tens of thousands of operations, and multi-step workflows ready for agents to use.
Does Jentic lock me in?
No, Jentic is open-standards-first (OpenAPI, Arazzo, MCP). Your API definitions, workflows, and integrations remain portable.
Can I bring my own private APIs?
Yes, you can import private APIs and workflows, then control access at the agent level. Self-hosted deployment support is on the roadmap.
Do I need Jentic to build a Business Agent?
No, you can build agents without Jentic, but it handles the undifferentiated heavy lifting like auth, cataloging, workflows, and observability aligned to open standards.
What is a Business Agent?
A Business Agent is an AI system that plans and executes real-world tasks by calling external APIs and workflows securely, observably, and under your organization's controls.
More in API Gateway
Kong GatewayKong is a self-hosted, open-source API and AI Gateway for proxying, routing, load balancing, auth, and LLM/MCP traffic with 60+ AI features.
krakendSelf-hosted, open-source API gateway: a single stateless binary that aggregates, secures and transforms microservice APIs, with Docker, deb/rpm and brew installs.
llm-gatewaySelf-hosted open-source API gateway routing 250+ LLM models from 40+ providers via unified OpenAI-compatible API with cost tracking, caching, and failover.
llm-gatewaySelf-hosted OpenAI-compatible API gateway routing to OpenAI, Anthropic, and local backends with semantic routing, load balancing, and zero-trust networking via zrok.
lmgateSelf-hosted gateway adding enterprise auth, rate limits, and model ACLs to Ollama, llama.cpp, and LM Studio via Docker Compose.
LumenLumen is a self-hosted AI gateway. It provides a web chat interface and an OpenAI-compatible API proxy/
LynkrSelf-hosted, open-source LLM gateway for AI coding assistants that compresses tool tokens, caches prompts semantically, and routes by complexity across 14+ providers.
ManifestSelf-hosted, open-source LLM gateway: one OpenAI-compatible endpoint routes agents to 300+ models with provider fallbacks, cost tracking and full request logs.
mcp-context-forgeAn AI Gateway, registry, and proxy that sits in front of any MCP, A2A, or REST/gRPC APIs, exposing a unified endpoint with centralized discovery, guardrails and management. Optimizes Agent & Tool calling, and supports plugins.
mcp-gatewaySelf-hosted MCP gateway converting REST APIs into secured MCP endpoints with multi-tenant isolation, API key/OAuth auth, tool policies, and audit logging.
MCPJungleSelf-hosted, open-source MCP gateway: register stdio and HTTP MCP servers once, serve them via one /mcp endpoint with tool groups and access control.
McpManagerSelf-hosted open-source MCP proxy aggregating multiple upstream servers into a single endpoint for AI tools like Claude Code and Cursor.
0 comments