wiff

# wiff

**Harness-agnostic, deterministic, resumable multi-agent workflows — written as plain JavaScript. Like `wf`, but wiff.**

[![npm](https://img.shields.io/npm/v/%40xxxoooxoxo%2Fwiff?label=npm&color=cb3837)](https://www.npmjs.com/package/@xxxoooxoxo/wiff) [![MCP Registry](https://img.shields.io/badge/MCP_Registry-io.github.xxxoooxoxo%2Fwiff-6b46c1)](https://registry.modelcontextprotocol.io/v0/servers?search=io.github.xxxoooxoxo/wiff) ![MIT License](https://img.shields.io/badge/license-MIT-blue) ![Node >= 22](https://img.shields.io/badge/node-%3E%3D22-brightgreen)

Fan a task out to a fleet of agents with a small script instead of a prayer. You write ordinary JavaScript with `agent()`, `parallel()`, and `pipeline()`; the runtime executes it in the background, journals every step, and — when a run dies halfway through — resumes it without re-paying for a single completed agent. The engine is a plain MCP server with durable on-disk state, so the same run can be started, watched, resumed, or cancelled from **any** MCP client — Codex, Claude Code, Cursor, or a cron job — while each child runs on Codex, Claude, Cursor, or Kimi.

<picture>
  <source media="(prefers-color-scheme: light)" srcset="docs/screenshots/run-light.png">
  <img alt="A live wiff run in the viewer: a three-phase accessibility audit with one inventory agent completed and three audit agents running in parallel, each with a live status line, a gantt timeline, and per-agent model/effort/sandbox/token badges." src="docs/screenshots/run-dark.png">
</picture>

```js
export const meta = {
  name: "audit",
  description: "Audit files in parallel, fix confirmed issues in isolation",
  phases: [{ title: "Audit" }, { title: "Fix" }],
};

phase("Audit");
const findings = await parallel(
  args.files.map((file) => () =>
    agent(`Audit ${file} for auth bugs`, {
      key: `audit:${file}`,          // stable key → free replay on resume
      sandbox: "read-only",
      schema: findingSchema,          // structured JSON output
    }),
  ),
);

phase("Fix");
return await parallel(
  findings.filter((f) => f.real).map((f) => () =>
    agent(`Fix: ${f.summary}`, {
      key: `fix:${f.file}`,
      agentType: "surgeon",           // persona from .codex/agents/surgeon.md
      isolation: "worktree",          // own git worktree — parallel writes can't collide
      sandbox: "workspace-write",
    }),
  ),
);
```

## Why

**There is no harness-agnostic workflow orchestration system.** Every coding harness has some
multi-agent story — Claude Code has its Workflow tool, Codex has subagents, Cursor has its own
agents — but each one is welded to its harness: its runs live and die with that app, its state is
invisible to everything else, and none of them can be driven from anywhere but their own chat
window. wiff pulls orchestration out of the harness: the engine is a plain MCP server with durable
on-disk state, so **any** MCP client — Codex, Claude Code, Cursor, a cron job — can start, watch,
resume, or cancel the same runs, and the orchestration itself is a script rather than a
conversation.

Ad-hoc multi-agent orchestration ("spawn some subagents for this") is also great until the run is
40 agents deep and something dies. Workflows-as-code give you:

- **Determinism** — the orchestration is a script, not vibes. No time, randomness, filesystem, or network inside workflow code; agents do the external work.
- **Resume, not retry** — every agent call is journaled with a stable key and an input hash. Kill the host, edit the script, resume the run: unchanged completed agents replay from cache instantly and for free. Agents that were interrupted **mid-turn** re-run with a digest of their previous attempt's transcript injected ("here's what you already did — continue"), and worktree agents inherit their partial checkout instead of starting over.

  <img alt="A resumed run: attempt 2, with the inventory agent and all three audit agents replayed from cache in 0ms and only the interrupted synthesis agent re-running." src="docs/screenshots/resume-dark.png">

  That screenshot is the feature: the host was killed mid-synthesis, and on resume the four finished agents came back from the journal in 0ms — only the interrupted one re-ran.
- **Fail-hard semantics** — a rejected agent fails the workflow loudly (`parallelSettled()` is the explicit opt-out). No silent `null`s masquerading as success.
- **Visible scheduling** — agents are journaled as queued before they acquire a runtime slot and
  running only when backend execution starts. Queue and execution durations stay separate, while
  owner heartbeats make a live-but-stalled workflow visible.
- **Isolation where it matters** — `isolation: "worktree"` gives each writing agent a fresh detached git worktree. Clean ones vanish; dirty ones are kept and listed on the run for you to inspect or merge.
- **Personas** — `agentType: "reviewer"` injects a markdown persona as the child's developer instructions, with frontmatter defaults for model/effort/sandbox.

## Install

**Codex** (plugin: MCP tools + the `$workflow` authoring skill):

```sh
codex plugin marketplace add https://github.com/xxxoooxoxo/wiff.git
codex plugin add wiff@wiff
```

**Claude Code** (plugin: MCP tools + skill):

```sh
claude plugin marketplace add xxxoooxoxo/wiff
claude plugin install wiff@wiff
```

**Anything else** — the server is on npm ([`@xxxoooxoxo/wiff`](https://www.npmjs.com/package/@xxxoooxoxo/wiff)) and the [official MCP Registry](https://registry.modelcontextprotocol.io/v0/servers?search=io.github.xxxoooxoxo/wiff) (`io.github.xxxoooxoxo/wiff`), so registry-aware clients can install it by name, and everything else runs it with npx:

```sh
npx -y @xxxoooxoxo/wiff        # stdio MCP server
```

Or from a local checkout:

```sh
git clone https://github.com/xxxoooxoxo/wiff.git
codex plugin marketplace add ./wiff
codex plugin add wiff@wiff
```

Then start a new Codex session and either invoke the bundled skill with `$workflow` or just ask: *"run this as a resumable workflow."*

Installing the plugin auto-approves its five workflow-controller tools so headless and desktop runs don't stop at an MCP approval prompt. Agent filesystem access is still governed per-call by `sandbox`.

## Using from other harnesses (Claude Code, Cursor, any MCP client)

The Codex *plugin* is just packaging. The engine underneath is a plain stdio MCP server, so any
MCP-speaking harness can orchestrate wiff workflows. The mental model: **both the orchestrator
and the workers are pluggable** — whoever drives, each `agent()` child runs on a backend chosen
from its model name: `gpt-*`/`o*` models run as native Codex threads via a local
`codex app-server`, `claude-*`/`opus`/`sonnet`/`haiku`/`fable` models run as headless `claude`
agents, `composer-*` models run through the official Cursor SDK (`@cursor/sdk`) in-process, and
`kimi-code/*` models run as headless `kimi` processes. A workflow can mix them freely
(`provider: "codex" | "claude" | "cursor" | "kimi"` overrides the inference, `WIFF_BACKEND`
sets the fallback for unrecognized models). On the Claude, Cursor, and Kimi backends,
`workspace-write` requires `isolation: "worktree"`; Kimi's `read-only` mode is advisory because
print mode auto-approves tools and has no OS sandbox.

Requirements on the machine, regardless of harness: Node >= 22, git if you use
`isolation: "worktree"`, and the runtime of whichever backend your agents use — Codex CLI
>= 0.144.6
and/or `claude` CLI installed and authenticated, `CURSOR_API_KEY` for Cursor agents, or the
`kimi` CLI configured with the requested full model alias (for example `kimi-code/k3`).

**Claude Code** — the plugin install above is the easy path. To wire just the server manually:

```sh
claude mcp add wiff -- npx -y @xxxoooxoxo/wiff
```

Tool calls go through Claude Code's own permission system; to skip per-call prompts, allow the
five tools in `.claude/settings.json`:

```json
{ "permissions": { "allow": [
  "mcp__wiff__workflow_start", "mcp__wiff__workflow_status",
  "mcp__wiff__workflow_wait", "mcp__wiff__workflow_cancel",
  "mcp__wiff__workflow_models"
] } }
```

**Cursor / Windsurf / Claude Desktop** — add the server to the client's `mcp.json`:

```json
{
  "mcpServers": {
    "wiff": { "command": "npx", "args": ["-y", "@xxxoooxoxo/wiff"] }
  }
}
```

Notes for non-Codex hosts:

- **State is shared.** Every harness reads and writes the same `~/.wiff/runs/`, so a run started
  from Codex can be watched, cancelled, or resumed from Claude Code (and vice versa), and the
  live viewer sees everything.
- **Bring the script contract into context.** The `$workflow` skill only auto-loads inside Codex.
  From other harnesses, point the model at
  [`plugins/wiff/skills/workflow/references/api.md`](plugins/wiff/skills/workflow/references/api.md)
  (or copy the skill into your harness's skill/rules directory, e.g. `.claude/skills/` or Cursor
  rules) so it authors valid scripts.
- **Personas** resolve from `<cwd>/.codex/agents/` then `~/.codex/agents/` on every harness; set
  `CODEX_WORKFLOW_AGENTS_DIR` in the server's env to point somewhere else (e.g. a shared
  `~/.claude/agents`).
- `workflow_start` requires an explicit absolute `cwd`, so the server's own working directory
  doesn't matter to results.

## For agents

If you are a coding agent — driving wiff over MCP or hacking on this repo — read [AGENTS.md](AGENTS.md). It covers the five workflow tools, the script-authoring rules that actually catch agents out (stable `key`s, thunks not promises, no I/O in workflow code, worktree isolation for concurrent writers), where run state lives on disk, and how to verify changes to the runtime. The full script contract is in [the API reference](plugins/wiff/skills/workflow/references/api.md).

## How it works

The plugin is an MCP server exposing five tools: `workflow_start`, `workflow_status`, `workflow_wait`, `workflow_cancel`, and `workflow_models`. A started workflow runs its script inside a locked-d