Docs menu

Introduction

The trust boundary: a receipt after, a gate before.

Agenko is the deterministic trust boundary for AI agents — open source, OTEL-native, Apache 2.0. One line of SDK gives every action a receipt it cannot forge and a gate it cannot cross.

The one rule: receipt after (@check) + gate before (@approve). Detection tools watch. Agenko proves what happened and stops the irreversible call.

Why Agenko exists

Most agent observability stacks help you watch production — traces, scores, prompt versions, issue detection. Useful, but watching is not proving, and it is not stopping.

LayerTypical observabilityAgenko
After the runTraces, scores, inferred issuesReceipt — claimed vs actual, hashed
Before the action—Gate — deterministic human approval
Signal typeOften LLM-inferredMeasured facts; estimated: true when not

If a signal is estimated (grounding score, future evals), it is labeled estimated: true. We never sell an inference as a measurement.

Who this is for

Teams running agents in production who need to:

  • Catch silent failures (200-but-wrong, claimed tool never ran)
  • Stop irreversible actions before they execute
  • Keep traces and receipts in their own cloud (self-host first)
  • Stay OTEL-native with Grafana, Tempo, or any collector you already run

How it works

  1. Install and init(). Zero required args. Reads AGENKO_API_KEY / AGENKO_HOST from env. Never throws.
  2. Wrap actions with @check. Every return gets _agenko_receipt: claimed, actual, verdict, hash chain.
  3. Gate irreversible work with @approve. SDK posts to the control plane; console inbox is the human side.
  4. Traces flow over OTLP. OpenInference attributes on every span. Console shows receipts, traces, approvals.
bash
pip install agenko
# or: pnpm add @agenko/sdk
python
import agenko

agenko.init()

@agenko.check
def handle_ticket(ticket):
    return process(ticket)

@agenko.approve(when="irreversible")
def delete_account(user_id):
    db.delete_user(user_id)

What ships today

PrimitiveStatus
init()Shipped — zero required args, never throws
@check receiptsShipped — Python + TypeScript
@approve gatesShipped — polls the control plane
Auto-instrumentationShipped — OpenAI, Anthropic, LangChain, LiteLLM, and more
Console (traces / receipts / approvals)Shipped against the control plane
why() graph walkNot shipped yet
Prompt registry (get_prompt, SWR cache, fallback)Shipped
Eval harnessNot shipped as a public API

Architecture

code
Your agent
   │
   ├─ OTLP traces ──► control plane (/api/public/otel/v1/traces)
   ├─ REST receipts / approvals ──► control plane (/api/v1/...)
   └─ Console ◄── Postgres + worker

Python and TypeScript export the same shape: init, check, approve, get_client.

Choose your path

GoalStart here
First receipt in 30 secondsQuickstart
Wire a coding agentAgent skill · Best practices
Understand the nounsCore concepts
Wire OpenAI / LangGraphIntegrations
Run in your VPCSelf-hosting
What we borrow vs buildOpen source reuse

For AI agents

Fetch the documentation index at /llms.txt. Install the agent skill before writing SDK code. Every page ends with Agent instructions and a canonical URL. Ground answers in shipped SDK exports only — mark estimated: true until a check is a hard fact.

Agent instructions

This page is part of the Agenko documentation at docs.agenko.app. Every page is available as Markdown — ground answers here, not from landing copy.

Querying these docs

Before writing SDK code

  • Install the agent skill or read Agent skill
  • Read Best practices — what a good receipt looks like
  • Trust boundary: @check after, @approve before
  • init() — zero required args, never throws
  • Mark estimated: true until a check is a hard fact
  • Shipped vs backlog: Introduction status table

Page: Introduction
HTML: https://docs.agenko.app/introduction
Markdown: https://docs.agenko.app/introduction.md