arthai-marketplace

What is arthai?

arthai is a development toolkit that turns Claude Code into a coordinated engineering team, specialized agents, slash-command workflows, and automatic guardrails, installed as Claude Code plugins and adapted to your codebase.

The problem it solves

Claude Code out of the box is one generalist assistant. It’s powerful, but:

• Everything runs at premium cost. A file lookup costs the same model tier as an architecture decision.
• Workflows live in your head. Plan → implement → QA → PR is a sequence you re-type and re-forget every time, and steps get skipped.
• No guardrails. Nothing stops a destructive command, an accidental edit to a migration file, or a push to main.
• No visibility. When a session ends, you have no record of what the AI did, which agents ran, or what it cost.

What arthai adds

Layer	What it is	Example
Agents	Specialists with the right model tier for the job: QA, SRE, frontend, backend, planning, research	A Haiku agent handles file searches at 1/60th the cost of Opus
Skills	Slash-command workflows that encode multi-step engineering sequences	/fix #42 runs root-cause analysis, scope lock, fix, regression tests
Hooks	Automatic behaviors at session lifecycle events	A guard blocks rm -rf and force-pushes before they execute
Calibration	/calibrate deep-learns your codebase and seeds a knowledge base; the knowledge graph builds from it automatically at session start and workflows query it	Fixes match your conventions; QA understands your domain rules
Observability	Arth Intelligence: a local dashboard of every session, tool call, agent spawn, and dollar	See exactly what the AI did and what it cost, on your machine

The three layers: what each adds

Each layer is optional and stacks on the previous one. Start with the toolkit; add the others when you want what they give you.

flowchart LR
    subgraph L1["1 · Toolkit (arthai plugins)"]
        A["38 agents · 61 skills · 29 hooks\n/calibrate knowledge base\nlexical knowledge-graph retrieval\ncheaper-model routing"]
    end
    subgraph L2["2 · + Arth Intelligence (OTEL)"]
        B["sessions / cost / tokens dashboard\nagent DAG & workflow story\n/experiments A-B comparisons\nExplain-this-session summaries"]
    end
    subgraph L3["3 · + arth-memory"]
        C["semantic knowledge retrieval\ncited, budget-packed context\nconstraint checks via MCP\n0-token calibrate on new clones"]
    end
    L1 --> L2 --> L3

Layer	You get	Cost of adding it
Toolkit alone	Skills, agents, hooks; /calibrate learns your codebase; workflows query the knowledge base lexically	License + plugin install, no services
+ Intelligence (experimental preview)	Every session visualized: tool calls, agent spawns, cost per model, workflow phases; A/B experiments; AI session summaries	One /otel-setup run, Docker, ~1.5GB
+ Memory	Same skills now retrieve knowledge semantically (meaning, not keywords), with citations and constraint checks; new clones hydrate the KB for 0 tokens	One /memory-setup run, Docker, ~800MB, no GPU

Retrieval is upgraded in place: the same kg-query entry point every workflow uses routes to memory when it’s on and falls back to the lexical graph when it isn’t. The skills themselves never change.

Who it’s for

• Engineers shipping with Claude Code daily who want repeatable workflows instead of improvised prompting.
• Teams that care about cost, the toolkit routes routine work to cheaper models automatically.
• Anyone who needs an audit trail of what AI did in their codebase.

It also includes role-based flows (/welcome, /wizard) for non-engineers on the team.

What it is NOT

• Not a replacement for Claude Code, it’s installed into Claude Code as plugins.
• Not cloud-required, the default install runs everything locally and observability data never leaves your machine. Optional cloud tiers exist per layer for teams (see Deployment Modes).
• Not all-or-nothing, install focused bundles (QA only, ops only, consulting only) or prime for everything.

How the pieces fit

flowchart LR
    You([You]) --> CC[Claude Code]
    CC --> P[arthai plugins]
    P --> TR["Triage router (hook)<br/>picks the cheapest capable route"]
    TR --> SK["Skills<br/>/planning → /implement → /qa → /pr"]
    SK --> AG["Agents<br/>tiered Haiku · Sonnet · Opus"]
    P -. OTEL hook .-> AI[("Arth Intelligence<br/>local Docker dashboard")]
    CC -. native OTEL: cost + tokens .-> AI

Next steps

• Getting started, install in ~10 minutes
• Plugin catalog to pick a bundle
• Arth Intelligence, turn on the dashboard
• FAQ