RYUKSAIDSOdocumentation
Start here

Overview

What Ryuksaidso is, and the words it uses

v2.0.0Updated Sep 25, 20263 headings

Ryuksaidso is an agent control plane — a single place to define AI agents, execute them as durable jobs, pause their risky side effects for a human decision, and keep the evidence that proves they behave. It is the tooling you wish you had the first time an agent did something “weird” in production and nobody could say why.

What you get

Durable traces

Every run persists its planner, tool and synthesis steps with inputs, outputs, latency and token usage. Nothing lives only in a log line.

Approval gates

Policies pause dangerous tool calls until a person approves or rejects. Approvals are scoped, timed and written to the audit trail.

Versioned agents

Draft freely, publish immutable versions with changelogs, pin one to production and roll back without redeploying anything.

Evaluations

Regression datasets score intent accuracy for every draft, so a version ships because it beat the last one — not because it felt right.

Knowledge grounding

Postgres full-text retrieval keeps answers anchored in your own documents, scoped to your workspace.

Multi-provider LLMs

Route between Ollama (local) and OmniRoute (cloud), switch models from the UI, and let the worker fail over automatically.

How the platform fits together

PieceTechnologyJob
WebNext.js 15 · React 19Control plane UI at :3000
APIExpress · Zod · PrismaREST endpoints, auth, rate limits at :4001
WorkerBullMQ on RedisExecutes runs, tools, approvals and evals
DatabasePostgreSQL (pgvector)Agents, runs, traces, knowledge, audit
Cache & queueRedis 7Sessions, rate limits, durable job queue
LLM providersOllama · OmniRouteModel inference for agents and evals

Vocabulary

TermMeaning
WorkspaceThe tenant boundary. Every row in the database is scoped to one.
ProjectA grouping for related agents, with its own production version pin.
AgentInstructions, a tool allowlist and a knowledge scope. Editable draft.
VersionAn immutable snapshot of an agent, published with a changelog.
RunOne execution of an agent version, queued as a durable job.
StepA single action inside a run — plan, tool call or synthesis.
PolicyA rule that decides whether an action needs human approval.
ApprovalThe pending decision a human must make before a run continues.
EvaluationA dataset run against an agent to score its behaviour.
New here?
Head to the Quick start and you will have a traced, approval-gated run in a few minutes. Everything else on this page assumes that baseline.
RYUKSAIDSO docs · built for people who ship agents for a living