jentic

Plug your AI agents into a vast open repository of API operations and workflows with a single turn-key integration, without ever handing over credentials.
Score and improve your OpenAPI documents for AI-readiness. Build, validate, and run Arazzo workflows.
All the capability, without the complexity. Open, secure, and reliable.

Documentation • Quickstart • Website

• Jentic One – (⚠️ Public Beta) An open-source self-hosted API execution layer that sits between your agent and the outside world. Your agent says what it wants to do. Jentic One handles the how — finding the right API, injecting credentials at runtime, and brokering the request.

  • Search, execute, observe, and manage API access with a local credentials vault and scoped toolkit keys per agent
  • Fully API-compatible with the hosted and VPC Jentic editions
  • Perfect for: self-hosters, privacy-conscious teams, local agent development

• Jentic Skills – Agent skills for the Jentic platform

  • jentic – Core Jentic skill for self-hosted Jentic Mini
  • jentic-v1 – Skill for hosted Jentic (jentic.com)
  • jentic-workflows – Generates Arazzo workflow YAML from a natural-language goal or agent tool trace
  • Perfect for: OpenClaw users, agent workflow automation

• Jentic SDK (Python) – Official Python SDK for discovering and executing APIs and workflows in your agents

  • Perfect for: Python developers, agent builders, tool integrators

• Jentic Tools – MCP server for connecting any MCP-compatible agent to Jentic

  • Perfect for: Claude Desktop, Cursor, Windsurf, ChatGPT, VS Code MCP users

• Jentic Public APIs – Open catalog of 6,000+ AI-ready API specs and 2,000+ Arazzo workflows

  • Perfect for: API documentation writers, knowledge contributors

• Awesome Agent APIs - A curated list of the most AI-agent-ready APIs, scored by the Jentic API AI-Readiness Framework

• Standard Agent – Composable Python agent library with Jentic tools built in

  • Perfect for: Python developers building reasoning agents

• API AI-Readiness Framework (JAIRF) – Open specification for evaluating how well an API supports AI agents and LLMs across six scored dimensions

  • Perfect for: API designers, platform architects, governance teams

• Jentic API Scorecard – Score and improve any OpenAPI document against our AI-Readiness framework and get a single grade with prioritised recommendations

  • CLI – npx @jentic/api-scorecard-cli score ./openapi.yaml — runs locally in Docker, spec never leaves your machine
  • GitHub Action – gate PRs on AI-readiness score, upload SARIF findings to the Security tab
  • Agent Skills – install into Claude Code, Vercel skills CLI, or TanStack Intent. Dedicated score and improve skills
  • Web UI – paste a URL or drop a file at jentic.com/scorecard
  • Perfect for: API developers, designers, product managers, platform teams, CI/CD pipelines

Give your agents well-formed workflows and API specs to work with. These libraries cover the full lifecycle — parse, resolve, validate, run, and render.

• Arazzo Engine – Python toolkit for executing and generating Arazzo workflow specifications

  • Arazzo Runner – Executes Arazzo workflows end-to-end from the CLI or Python
  • Arazzo Generator – Generates Arazzo workflows from OpenAPI Descriptions using AI
  • Perfect for: Python developers, workflow automation, CI pipelines

• Jentic Arazzo Tools – TypeScript/JavaScript monorepo for parsing, resolving, validating, running, and rendering Arazzo documents

  • @jentic/arazzo-parser – Parses Arazzo documents into an ApiDOM data model
  • @jentic/arazzo-resolver – Dereferences Arazzo and OpenAPI documents
  • @jentic/arazzo-validator – Validates and lints Arazzo documents via CLI or API
  • @jentic/arazzo-runner – Executes Arazzo workflows in TypeScript/JavaScript environments
  • @jentic/arazzo-ui – UI components for rendering Arazzo workflows
  • Perfect for: TypeScript/JavaScript developers, tooling and UI builders

• Jentic OpenAPI Tools – Python toolkit for working with OpenAPI documents

  • Perfect for: Python developers, API tooling builders

• API Problem Details – Reusable RFC 9457 Problem Details components for consistent, machine-readable error responses across all Jentic APIs

  • OpenAPI schemas — reference-ready response definitions for all standard error codes
  • jentic-problem-details – Python package with Pydantic models and FastAPI utilities
  • @jentic/problem-details – TypeScript type definitions and utilities
  • Perfect for: API developers building agent-consumable error responses, FastAPI and TypeScript projects

• Jentic on YouTube – Walkthroughs, demos, and agent builder tutorials

We welcome contributors of all backgrounds and experience levels!

• Start with issues labeled good first issue in any of our repositories
• Check out Standard Agent's contribution guide
• Join our Discord for friendly help and guidance

• Look for enhancement and feature labeled issues
• Propose new features or architectural improvements
• Help review pull requests from new contributors

• Improve API documentation and examples
• Create tutorials and guides
• Fix typos and clarify explanations

• Help improve our documentation design
• Create diagrams and visual aids
• Enhance user experience across our tools

Connect with agent builders, API developers, and workflow automation experts:

• Weekly Ask Me Anything: Join our Discord for live Q&A sessions
• Contributor Spotlights: We feature outstanding contributors monthly
• Agent Builder Meetups: Connect with other developers building AI agents
• API Workshop Series: Learn about integrating APIs with AI agents

We believe in recognizing our contributors:

• Contributor Hall of Fame in our documentation
• Special Discord roles for active contributors
• Early access to new features and tools
• Jentic swag for significant contributions
• Speaking opportunities at our community events

🧩 Jentic is built for developers who want to move fast, skip boilerplate, and give their agents real-world capabilities without complexity.

Ready to contribute? Start by exploring our repositories above, joining our Discord, and finding your first issue to tackle!