agentjido

agentjido / jido_managed_agents

Public

Jido implementation of Managed Agents on Phoenix

jido.run
17
2
89% credibility
Found Apr 12, 2026 at 17 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Elixir
AI Summary

Self-hosted platform for creating, running, and managing AI agents with a dashboard, API compatibility, and durable session traces.

How It Works

1
🔍 Discover open agent manager

You find this free project as a self-hosted way to run and manage AI helpers like those from big services.

2
🚀 Start it on your computer

Follow simple steps to get the app running locally so you can use it right away in your browser.

3
👋 Sign in and explore demos

Log in with the ready example account to see sample agents and past conversations instantly.

4
🔑 Make your private access code

Create a secure personal code to unlock full features for your own scripts and tools.

5
🛠️ Build or add an AI helper

Define your agent's smarts, tools, and safe settings using easy examples or your ideas.

6
▶️ Chat with your agent

Start a session to watch it think, use tools, connect services, and reply step by step.

📊 Review every detail

Check the dashboard traces to see full histories, tool actions, and results of your agent's work.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 17 to 17 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is jido_managed_agents?

This Elixir Phoenix app delivers a self-hosted managed agents platform, implementing Anthropic-compatible APIs and workflows with Jido's supervised runtime. Developers get a local stack for building, running, and tracing AI agents: import YAML definitions, configure cloud-like environments and vaults for credentials, and execute sessions via `/v1` endpoints or a dashboard at `/console`. It solves the need for an open, inspectable alternative to hosted services, seeding demo agents and sessions out of the box after `mix setup` and `mix phx.server`.

Why is it gaining traction?

It drops a working end-to-end prototype days after Anthropic's announcement, letting devs run agents locally without vendor lock-in or costs. The `/v1` API uses `x-api-key` auth and mirrors Anthropic envelopes, while the dashboard offers inline runs, traces, and resource management—plus MCP tools at `/mcp`. Examples and scripts make experimentation instant, fueling community PRs on this pre-alpha base.

Who should use this?

Elixir/Phoenix teams prototyping agentic apps or migrating from hosted LLMs. Jido users extending to multi-agent sessions with vaults and workspaces. Devs auditing AI runs need the event logs and threaded traces for debugging tool calls and delegations.

Verdict

Grab it for local agent tinkering if you're in the Elixir/Jido ecosystem—quickstart shines despite 17 stars and pre-alpha rough edges. Credibility score of 0.9% reflects early days, but seeded demos and open architecture invite contributions to harden it fast. (187 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.