🟪 Open-source runtime that ships any LangGraph or Google ADK agent as a production-ready FastAPI service. Bundled , AG-UI copilotkit API, chat UI, 15+ guardrails, MCP, OpenTelemetry, OIDC. One pip install. Self-hosted, no vendor lock-in.
What's inside
Drive Google's Antigravity CLI (agy, powered by Gemini 3.5) without leaving Claude Code. A Claude Code plugin that hands work to agy — Google's Antigravity CLI — and brings the result back into your session.
@idun-groupFAQ
idun-agent-platform is a Claude Code plugin with hand-picked skills for deployment work, indexed on Flowy. Install it with the command on its page. Its skills do not fire on their own yet. Request auto-invocation to have Flowy route them as you prompt. Free and open source.
Repo: Idun-Group/idun-agent-platform
Self-hosted. Open source. No vendor lock-in.
Cloud · Try the demos · Quickstart · Docs · Discord · Book a demo
⭐ If you find this useful, please star the repo. It helps others discover the project.
Why Idun? Teams building agents face a tradeoff: build the production wrapper yourself (FastAPI + traces + guardrails + admin UI — slow), or adopt a SaaS like LangGraph Cloud or LangSmith (sovereignty trade-off). Idun is the third path: pip install a self-sufficient FastAPI process that bundles your agent with chat UI, admin, traces, and guardrails — all open source, all on your infrastructure.
Prerequisites: Python 3.12 or 3.13.
Create a new directory and install the engine:
mkdir my-agent && cd my-agent
pip install idun-agent-engine langgraph langchain-google-genai
Save the three files below inside my-agent/.
my-agent/agent.py
from typing import Annotated, TypedDict
from idun_agent_engine.mcp import get_langchain_tools_sync
from langchain_google_genai import ChatGoogleGenerativeAI
from langgraph.graph import StateGraph, START
from langgraph.graph.message import add_messages
from langgraph.prebuilt import ToolNode, tools_condition
class State(TypedDict):
messages: Annotated[list, add_messages]
tools = get_langchain_tools_sync()
llm = ChatGoogleGenerativeAI(model="gemini-2.0-flash").bind_tools(tools)
def chatbot(state: State):
return {"messages": [llm.invoke(state["messages"])]}
graph = StateGraph(State)
graph.add_node("chatbot", chatbot)
graph.add_node("tools", ToolNode(tools))
graph.add_edge(START, "chatbot")
graph.add_conditional_edges("chatbot", tools_condition)
graph.add_edge("tools", "chatbot")
my-agent/config.yaml
server:
api:
port: 8000
agent:
type: LANGGRAPH
config:
name: "my-agent"
graph_definition: "./agent.py:graph"
checkpointer:
type: sqlite
db_url: "sqlite:///conversations.db"
mcp_servers:
- name: idun-docs
transport: streamable_http
url: https://docs.idun-group.com/mcp
my-agent/.env — get a Gemini key at aistudio.google.com/apikey:
GEMINI_API_KEY=your-key-here
From inside my-agent/, run:
idun init
graph_definition: "./agent.py:graph"is resolved against the directory where you runidun init. Stay insidemy-agent/when you launch it, or use an absolute path.
A browser opens at http://localhost:8000 for the chat UI. The agent already has tools from the Idun docs MCP wired in. Visit /admin to configure more MCP servers, managed prompts, guardrails, observability, messaging integrations, and SSO. Visit /admin/traces for the trace store.
What
pip install idun-agent-enginedeliversA single wheel that bundles three Python packages:
idun_agent_engine— the engine SDK (FastAPI app factory, MCP registry, observability, guardrails).idun_agent_standalone— theidunCLI, the admin REST API, and the chat / admin / traces UIs.idun_agent_schema— the Pydantic config schemas.The
iduncommand on your$PATH(setup,serve,init,hash-password,agent serve) is provided by the bundled standalone — it is not a separately published package. You won't seeidun-agent-standalonein the engine's declared PyPI dependencies because it ships co-installed inside the wheel, not as a transitive dep.See docs.idun-group.com/architecture for the full layering.
Langfuse · Arize Phoenix · LangSmith · GCP Trace · GCP Logging
Trace every agent run. Connect multiple providers at the same time through config.
PII detection · Toxic language · Ban lists · Topic restriction · Bias checks · NSFW · 9 more
Apply policies per agent on input, output, or both. Powered by Guardrails AI.
Register MCP servers and control which tools each agent can access. Supports stdio, SSE, streamable HTTP, and WebSocket.
PostgreSQL · SQLite · In-memory · Vertex AI · ADK Database
Conversations persist across restarts. Pick a backend per agent.
Versioned templates with Jinja2 variables. Assign prompts to agents from the UI or API.
[!NOTE] AG-UI streaming — Every agent gets a standards-based streaming API, compatible with CopilotKit clients. Built-in chat playground for testing.
Idun ships as one process: idun-agent-standalone. It bundles the engine SDK, a Next.js chat UI, an admin panel, and a traces viewer — your agent runs inside this process, configured from a YAML file and re-loaded live from the admin REST.
flowchart LR
subgraph Idun["idun-agent-standalone (one process)"]
direction TB
UI["Chat UI / Admin / Traces"] --> ENG["Engine SDK"]
ENG --> DB[(Postgres / SQLite)]
end
Users --> UI
Admin --> UI
ENG --> Agent["Your LangGraph / ADK agent"]
Agent --> LLM["LLMs / MCP / tools"]
[!NOTE] Framework support — LangGraph and Google ADK are first-class today, with full adapters in the engine. LangChain is supported via the LangGraph adapter; broader native LangChain compatibility is on the roadmap.
| Idun Engine | LangGraph Cloud | LangSmith | DIY (FastAPI + glue) | |
|---|---|---|---|---|
| Self-hosted / on-prem | ✅ | ❌ | ❌ | ✅ |
| Multi-framework (LangGraph + ADK) | ✅ | LangGraph only | ❌ obs only | Manual |
| Guardrails (15+ built-in) | ✅ | ❌ | ❌ | Build yourself |
| MCP tool governance | ✅ per-agent | ❌ | ❌ | Build yourself |
| Observability (multi-provider) | ✅ Langfuse, Phoenix, LangSmith, GCP | ❌ LangSmith only | ✅ LangSmith only | Manual |
| Memory / checkpointing | ✅ Postgres, SQLite, in-memory | ✅ | ❌ | Build yourself |
| AG-UI / CopilotKit streaming | ✅ | ✅ | ❌ | Manual |
| Open source (GPLv3) | ✅ | ❌ | ❌ | — |
[!NOTE] Idun is not a replacement for LangSmith (observability) or LangGraph Cloud (hosting). It is the layer between your agent code and production that handles governance, security, and operations, regardless of which observability or hosting you choose.
After pip install idun-agent-engine, your $PATH and site-packages contain:
| Module | Distributed via | Purpose |
|---|---|---|
idun_agent_engine | declared in requires_dist | FastAPI app factory, MCP registry, observability, guardrails wiring |
idun_agent_schema | declared in requires_dist | Pydantic config models for config.yaml |
idun_agent_standalone | co-bundled in the engine wheel | idun CLI, admin REST under /admin/api/v1/, chat / admin / traces UIs at /, /admin/, /admin/traces/ |
idun (console script) | [project.scripts] idun = "idun_agent_standalone.cli:main" | Subcommands: setup, serve, init, hash-password, agent serve |
The bundled-install model means you do not need to pip install idun-agent-standalone separately — it isn't published. To verify your install carries everything:
python -c "import idun_agent_engine, idun_agent_standalone, idun_agent_schema; print('all 3 modules importable')"
idun --help
Both checks are smoke-tested in the release pipeline before any wheel is published.
Heads up: both pypi.org and test.pypi.org JSON APIs report entry_points: null and omit idun-agent-standalone from requires_dist. This is a Warehouse rendering quirk — the wheel itself still wires the idun console script and bundles standalone. The smoke-test commands above are the canonical way to verify a healthy install.
Every agent is configured through a single YAML file. Here is a complete example with all features enabled:
server:
api:
port: 8001
agent:
type: "LANGGRAPH"
config:
name: "Support Agent"
graph_definition: "./agent.py:graph"
checkpointer:
type: "sqlite"
db_url: "sqlite:///checkpoints.db"
observability:
- provider: "LANGFUSE"
enabled: true
config:
host: "https://cloud.langfuse.com"
public_key: "${LANGFUSE_PUBLIC_KEY}"
secret_key: "${LANGFUSE_SECRET_KEY}"
guardrails:
input:
- config_id: "DETECT_PII"
on_fail: "reject"
reject_message: "Request contains personal information."
output:
- config_id: "TOXIC_LANGUAGE"
on_fail: "reject"
mcp_servers:
- name: "time"
transport: "stdio"
command: "docker"
args: ["run", "-i", "--rm", "mcp/time"]
prompts:
- prompt_id: "system-prompt"
version: 1
content: "You are a support agent for {{ company_name }}."
tags: ["latest"]
[!TIP] Environment variables like
${LANGFUSE_SECRET_KEY}are resolved at startup. You can use.envfiles or inject them through Docker/Kubernetes.
Serve the engine directly from a YAML file (no admin UI, just the agent endpoints):
idun agent serve --source file --path config.yaml
[!IMPORTANT] Full config reference: docs.idun-group.com/configuration
9 runnable agent examples: idun-agent-template
| Questions and help | Discord |
| Feature requests | GitHub Discussions |
| Bug reports | GitHub Issues |
| Contributing | CONTRIBUTING.md |
| Roadmap | ROADMAP.md |
Maintained by Idun Group. We help with platform architecture, deployment, and IdP/compliance integration. Book a call · contact@idun-group.com
Minimal, anonymous usage metrics + masked session replay via PostHog. No message content, no PII beyond email after sign-in (set IDUN_TELEMETRY_IDENTIFY_USERS=false to disable). View source. Off: IDUN_TELEMETRY_ENABLED=false. Replay-only off: IDUN_TELEMETRY_SESSION_REPLAY=false.
.claude/
agents/
code-reviewer.md
guidelines/
rules/
MIGRATION-001.yaml
SCHEMA-001.yaml
SCHEMA-002.yaml
TEST-004.yaml
launch.json
.coderabbit.yaml
.coveragerc
.cursor/
rules/
engine.mdc
langgraph.mdc
project-overview.mdc
schema.mdc
.cursorignore
.cursorindexingignore
.dockerignore
.editorconfig
.env.example
.gitguardian.yaml
.github/
dependabot.yml
ISSUE_TEMPLATE/
bug_report.yml
config.yml
documentation.yml
feature_request.yml
PULL_REQUEST_TEMPLATE.md
workflows/
ci.yml
e2e-real-llm.yml
publish_engine_testpypi.yml
publish_schema_testpypi.yml
release.yml
socket-scan-reusable.yml
socket-security.yml
standalone-ci.yml
.gitignore
.gitleaksignore
.pre-commit-config.yaml
.specstory/
.gitignore
.vscode/
launch.json
settings.json
tasks.json
bruno/
idun_agent_engine/
agent-run.bru
bruno.json
Copilotkit stream.bru
get config.bru
Invoke.bru
Stream.bru
CLAUDE.md
CODE_OF_CONDUCT.md
commitlint.config.js
CONTRIBUTING.md
docker/
cloud-run.example.yaml
Dockerfile.base
Dockerfile.example
docs/
.mintignore
.mintlify/
Assistant.md
.nvmrc
adr/
_template.md
0001-record-architectural-decisions.md
README.md
advanced/
sdk.mdx
api-reference/
introduction.mdx
apple-touch-icon.png
architecture.mdx
auth/
overview.mdx
sso.mdx
changelog/
overview.mdx
CLAUDE.md
cli/
overview.mdx
configuration.mdx
deployment/
hardening.mdx
overview.mdx
docs/
docs.json
superpowers/
plans/
2026-03-22-ui-integrated-docs.md
specs/
2026-03-22-ui-exploration-report.md
2026-03-22-ui-integrated-docs-design.md
enterprise/
audit-logs.mdx
multi-agent.mdx
overview.mdx
rbac.mdx
sso.mdx
faq.mdx
favicon.ico
favicon.svg
frameworks/
adk.mdx
custom-adapter.mdx
langgraph.mdx
overview.mdx
glossary.mdx
guardrails/
overview.mdx
reference.mdx
guides/
connect-your-agent.mdx
deepagents-with-idun.mdx
idun-assistant-copilot.mdx
langgraph-production-deployment.mdx
overview.mdx
programmatic-chat.mdx
images/
guides/
01-login.png
01-signup.png
02-agent-dashboard.png
03-create-basics.png
04-create-framework.png
05-create-enrollment.png
06-agent-overview.png
07-guardrails-catalog.png
08-observability-catalog.png
09-connect-to-platform.png
10-agent-connected.png
11-chat-conversation.png
12-create-pii-guardrail.png
13-pii-guardrail-created.png
14-create-langfuse.png
15-attach-observability.png
16-attach-guardrail.png
17-agent-fully-configured.png
18-guardrail-blocked.png
19-langfuse-trace.png
deepagents/
01-agent-config.png
02-agent-graph.png
03-chat-reasoning.png
04-chat-answer.png
05-memory-sqlite.png
06-chat-history.png
07-dashboard.png
08-traces-list.png
09-trace-tree.png
10-trace-waterfall.png
11-langfuse-config.png
12-langfuse-cloud.png
13-mcp-google-workspace.png
14-mcp-tools-discovered.png
15-send-mail-chat.png
16-send-mail-inbox.png
idun-assistant/
admin-langfuse.png
admin-mcp-tools.png
admin-mcp.png
admin-memory.png
admin-prompts.png
admin-sso.png
chat-demo.png
email-result.png
trace-detail.png
trace-waterfall.png
logos/
A2A.png
ag-ui.png
agent-development-kit.png
google-cloud.png
guardrails-ai.png
langchain-color.png
langfuse-color.png
langgraph-color.png
langsmith-color.png
mcp.png
phoenix.svg
Postgresql_elephant.png
readme/
agent-detail.png
banner.png
chat-mcp.png
chat.png
dashboard.png
demo.webp
guardrails.png
integrations.png
login.png
mcp.png
memory.png
observability.png
prompts.png
welcome-ui.png
ui/
admin-agent-config.png
admin-dashboard.png
admin-guardrails-add.png
admin-guardrails.png
admin-integrations-discord.png
admin-integrations-slack.png
admin-integrations-teams.png
admin-memory.png
admin-observability-gcp-logging.png
admin-observability-langfuse.png
admin-observability-langsmith.png
admin-observability-phoenix.png
admin-prompts-new.png
admin-prompts.png
admin-sso.png
admin-trace-detail-tree.png
admin-trace-detail-waterfall.png
admin-traces-list.png
agents-create-framework.png
agents-create-resources.png
agents-create-step1.png
agents-create-step2.png
agents-detail-api.png
agents-detail-chat.png
agents-detail-config.png
agents-detail-gateway.png
agents-detail-logs.png
agents-detail-overview.png
agents-detail-prompts.png
agents-list.png
chat-conversation.png
chat-sso-login.png
chat-welcome.png
guardrails-catalog.png
guardrails-create-banlist.png
guardrails-create-form.png
guardrails-list.png
integrations-catalog.png
integrations-create-discord.png
integrations-create-form.png
integrations-discord-form.png
integrations-list.png
integrations-slack-form.png
integrations-whatsapp-form.png
login.png
mcp-catalog.png
mcp-create-form.png
mcp-create-stdio.png
mcp-list.png
mcp-overview.png
mcp-tools-discovery-small.png
mcp-tools-discovery.png
memory-catalog.png
memory-create-form.png
memory-create-postgres.png
memory-list.png
observability-catalog.png
observability-create-form.png
observability-create-langfuse.png
observability-langfuse-form.png
observability-list.png
onboarding-workspace.png
prompts-create-form.png
prompts-list.png
settings-appearance.png
settings-page.png
settings-security.png
sso-create-form.png
sso-list.png
sso-page.png
users-list.png
index.mdx
integrations/
discord.mdx
google-chat.mdx
slack.mdx
teams.mdx
whatsapp.mdx
introduction.mdx
logo/
dark.svg
light.svg
mcp-servers/
docker-toolkit.mdx
overview.mdx
memory/
adk.mdx
langgraph.mdx
overview.mdx
observability/
arize-phoenix.mdx
custom-handler.mdx
gcp-logging.mdx
gcp-trace.mdx
langfuse.mdx
langsmith.mdx
overview.mdx
telemetry-events.mdx
traces.mdx
prompts.mdx
quickstart.mdx
readme/
README.ar.md
README.es.md
README.fr.md
README.zh.md
roadmap.mdx
scripts/
capture-screenshots.ts
package-lock.json
package.json
playwright.config.ts
test-results/
.last-run.json
snippets/
config-note.mdx
contact-cta.mdx
engine-install.mdx
prerequisites.mdx
standalone/
admin.mdx
cloud-run.mdx
customizing-ui.mdx
openapi.json
overview.mdx
styles.css
team/
CODING-GUIDELINES.md
CONTRIBUTING-FOR-CLAUDE-USERS.md
templates.mdx
troubleshooting.mdx
videos/
idun-showcase.mp4
examples/
guardrails/
adk/
.env.example
.gitignore
agent.py
config.yaml
pyproject.toml
README.md
langgraph/
.env.example
.gitignore
agent.py
config.yaml
pyproject.toml
README.md
integrations/
adk/
.env.example
.gitignore
agent.py
config.yaml
pyproject.toml
README.md
langgraph/
.env.example
.gitignore
agent.py
config.yaml
pyproject.toml
README.md
mcp/
adk/
.env.example
.gitignore
agent.py
config.yaml
pyproject.toml
README.md
langgraph/
.env.example
.gitignore
agent.py
config.yaml
pyproject.toml
README.md
memory/
adk/
.env.example
.gitignore
agent.py
config.yaml
pyproject.toml
README.md
langgraph/
.env.example
.gitignore
agent.py
config.yaml
pyproject.toml
README.md
observability/
... 1077 more© 2026 Flowy · Free and open source
Built for Claude Code · Not affiliated with Anthropic