disp8ch

One-line install

The one-line installers are the easiest path for non-technical users. They download a managed Node.js 22 runtime if needed, use Corepack or npx pnpm, fetch the app source, install dependencies, create a clean local workspace, start disp8ch, and open onboarding.

First Model Setup

The easiest path is /onboarding. disp8ch supports four setup paths:

For API key or local setup:

1. Choose Online and add an API key, or choose Local.
2. For Local, select Check this PC to inspect installed models, available RAM and VRAM, and detected runtimes.
3. Run the recommended Ollama or llama-server command, then select Use this setup.
4. Run validation, then open WebChat and send a message.

For Claude account OAuth:

1. Install and sign in to Claude Code on the same Windows user that runs disp8ch.
2. Keep the Claude Code credential file private. Do not copy .claude, OAuth token files, or auth JSON into the repo.
3. In disp8ch, select or add an Anthropic model in Settings -> Models.
4. If the model form asks for a credential, use an environment or secret reference such as env:ANTHROPIC_TOKEN, env:ANTHROPIC_OAUTH_TOKEN, env:CLAUDE_CODE_OAUTH_TOKEN, or secret:CLAUDE_CODE_OAUTH_TOKEN.
5. Run the model test before using the model in WebChat or workflows.

For Codex account sign-in:

1. Install the Codex CLI and sign in locally with your Codex account.
2. Keep Codex auth files outside the repo and outside .env.local.
3. Leave normal WebChat on your selected provider or local model. Codex sign-in is only used when you explicitly choose the Codex coding-agent backend for delegated coding work.
4. Test with a harmless read-only delegation before granting write access.

You can also configure .env.local:

cp .env.example .env.local

Direct provider examples:

OPENAI_API_KEY=...
ANTHROPIC_API_KEY=...
ANTHROPIC_TOKEN=...
ANTHROPIC_OAUTH_TOKEN=...
CLAUDE_CODE_OAUTH_TOKEN=...
GOOGLE_API_KEY=...
DEEPSEEK_API_KEY=...

OpenRouter:

OPENROUTER_API_KEY=...

Local model endpoints:

OLLAMA_BASE_URL=http://127.0.0.1:11434
VLLM_BASE_URL=http://127.0.0.1:8000/v1
SGLANG_BASE_URL=http://127.0.0.1:30000/v1

Do not commit .env.local, .claude, .codex, auth JSON, OAuth token files, or any local credential store.

Quick Start

WebChat Examples

What can this app currently do? Separate implemented, configured, and callable.
List my automations and show which webhooks are enabled.
Create a webhook workflow that validates a GitHub-style JSON payload and summarizes it.
Compare three local model runtimes for an 8 GB VRAM laptop. Use current sources.
Audit this repo's API-key handling and cite exact files.
Create a board task for each blocker in this launch document.
Start a council session on whether we should prioritize reliability or new features.
Build a daily 9 AM research digest workflow, but ask before saving if anything is ambiguous.
Spin up a research team, put them in an org, and give them a board task to compare OCR models.
Generate a landing page concept for a local-first AI workspace and save it as a design.
Remember that I prefer concise technical answers. Reply only saved.
What is my preferred answer style?

Screenshots

<p align="center"> <img src="docs/readme-assets/agentic-workspace-loop.svg" alt="disp8ch agentic workspace loop" width="100%" /> </p>

One operating loop — Data Sources, WebChat, Council, Hierarchy, Workflows, Boards, Memory, Skills, Design Studio, Usage, and local model routing share the same workspace instead of acting like separate apps.

<p align="center"> <img src="docs/readme-assets/research-to-action.svg" alt="disp8ch research to action flow" width="49%" /> <img src="docs/readme-assets/automation-engine.svg" alt="disp8ch visual automation engine" width="49%" /> </p>

Research becomes work — source material can become cited answers, tasks, council sessions, workflows, and design artifacts. Automation stays visible — triggers, typed nodes, queues, traces, replay, and webhook responses are first-class runtime pieces.

<p align="center"> <img src="docs/readme-assets/dashboard.png" alt="disp8ch dashboard" width="100%" /> </p>

Dashboard — live system health, active workflows, agents, board tasks, execution lanes, and quick actions in one operator view.

<p align="center"> <img src="docs/readme-assets/webchat.png" alt="disp8ch webchat" width="49%" /> <img src="docs/readme-assets/workflows.png" alt="disp8ch workflow canvas with connected template nodes" width="49%" /> </p>

WebChat is the plain-English control surface for asking questions, inspecting app state, creating tasks, and running agentic tool work. Workflows is the visual automation canvas shown with real connected nodes: trigger → org context → agent brief → council/board follow-up → WebChat output.

<p align="center"> <img src="docs/readme-assets/hierarchy.png" alt="disp8ch full agent organization hierarchy" width="100%" /> </p>

Hierarchy shows the whole agent organization together: roles, goals, reporting lines, heartbeats, governance context, budget status, workload, and agent ownership. Other major surfaces include Boards for task flow, Council for structured debate, Data Sources for searchable context, Skills/Extensions/MCP for tool growth, Automations for cron and webhooks, and Design Studio for generated artifacts.

<p align="center"> <img src="docs/readme-assets/agent-ops-control.svg" alt="disp8ch agent operations control plane" width="100%" /> </p>

Option A — Ollama (easiest)

1. Install Ollama and start it. 2. In onboarding, select Check this PC and use the exact ollama run ... command shown for the recommended model. Ollama downloads that model only after you run the command yourself:

ollama serve
ollama run <recommended-model-tag>

1. Open onboarding at http://localhost:3100/onboarding, choose Local, select Check this PC, run the shown command, then select Use this setup, test, and save. No key required.

Memory search works without choosing a separate provider. New installs default to disp8ch's built-in local embedding model (Xenova/all-MiniLM-L6-v2) and fall back to keyword search if the model cache is unavailable. If you prefer Ollama embeddings instead, run ollama pull nomic-embed-text, set Settings -> Memory -> Embedding model to nomic-embed-text, then click Rebuild Index.

Or via .env.local:

OLLAMA_BASE_URL=http://127.0.0.1:11434

Option B — LM Studio, llama.cpp, vLLM, or SGLang (OpenAI-compatible)

1. Start your local server and load a model.
2. In onboarding choose the LM Studio / OpenAI-compatible preset and set the base URL. Leave the API key blank when the local server does not require one:

1. Run the test and save.

Or via .env.local:

VLLM_BASE_URL=http://127.0.0.1:8000/v1
SGLANG_BASE_URL=http://127.0.0.1:30000/v1

Tip: do not choose from parameter count alone. Context size, quantization, architecture, current free RAM/VRAM, and runtime support all affect whether a model is practical. Real AI image generation, live web search providers, external channels, and third-party APIs still need their own credentials, but the core local workspace runs without a model-provider key.

Run Fully Local (No API Key)

You do not need a cloud account or API key for core use. Run a local model server and point disp8ch at it — chat, local tools, memory, workflows, agents, boards, council, local document research, and local artifact work can run without a model-provider key. Live web search, external channels, cloud image generation, and third-party APIs still need network access and the credentials you choose to configure.

<p align="center"> <img src="docs/readme-assets/local-model-stack.svg" alt="disp8ch local model stack" width="100%" /> </p>

Useful CLI Commands

pnpm dpc status
pnpm dpc health
pnpm dpc doctor
pnpm dpc models list
pnpm dpc workflows list
pnpm dpc boards list
pnpm dpc orgs list
pnpm dpc skills list
pnpm dpc backup status
pnpm dpc learning status
pnpm dpc goals list

Developer checks:

pnpm install:test
pnpm exec tsc --noEmit
pnpm build

Desktop checks:

pnpm desktop:build
pnpm desktop:smoke
pnpm desktop:installer-smoke

Visual Workflow Automation

• Drag-and-drop node canvas for message triggers, webhooks, cron, manual triggers, GitHub events, agent calls, HTTP, RSS feeds, files, documents, memory, logic, boards, channels (incl. SMS and GitHub comments), and utility actions.
• Workflow templates for chat assistants, task routing, monitoring, scheduled reports, data processing, document intelligence, docs-site crawling, RSS/news monitoring, local lead enrichment, support/community triage with human-review drafts, evidence-backed strategy hardening loops, research loops, experiment loops, code review, channel intake, ops control towers, crew orchestration, short-video/content pipelines, and integrations.
• A ready-made automation recipe pack: nightly issue triage, pull-request review, docs-drift detection, dependency vulnerability scanning, deploy smoke verification, incident alert correlation, endpoint uptime watch, competitor-repo watching, weekly news digests, and a research-paper scanner — each pre-wired with a trigger, agent, and delivery.
• Notify-only-on-change: an agent (or node) can emit [SILENT] and the downstream send/notification node suppresses delivery — so scheduled checks stay quiet until something actually needs attention.
• Import/export, duplicate, replay, node testing, run-to-node, versions, trace drawer, credentials, data mapping, expression preview, and workflow-as-agent-tool behavior.
• An Executions view across all workflows with status filters, retry, and retry-from-failed-node.
• Per-workflow concurrency control: skip duplicate starts (default) or queue them durably (FIFO) with a max-concurrent limit — queued starts survive restarts.
• Per-workflow budget and escalation policy: cap runs/cost per day with optional auto-disable, and route threshold/failure escalations with notification limits and quiet hours — so unattended automations stay within guardrails.
• Webhook-triggered workflows can answer the HTTP caller directly with a response node (custom status/body/headers), or return a poll URL if the run takes longer.
• Import workflow JSON from other visual automation tools, with unsupported nodes preserved as visible placeholders instead of silently discarded.

Connect Anything: MCP, Extensions, And Tools

• Open MCP Servers under Capabilities to connect Model Context Protocol tools and resources. The catalog, per-tool enablement, approval policy, connection diagnostics, and named agent access picker live on this dedicated page. Skills and Extensions remain separate because they are reusable capability packs, while MCP configures external processes, credentials, trust, and runtime access.
• Install extension packs for channels, providers, memory backends, and integrations, and enable them per agent. In Hierarchy → Ops, an approved preset can be merged into every current organization member without removing existing capabilities.
• Define custom tools and expose any workflow as an agent tool — extend the agent without forking the app.

FAQ

Do I need an API key or a cloud account? No for core local use. disp8ch can run with Ollama, LM Studio, llama.cpp, vLLM, or SGLang — see Run Fully Local. Cloud providers and OpenRouter are optional. Claude account OAuth is supported for Anthropic model access when you already use Claude Code. Codex sign-in is supported for optional coding-agent delegation, not as the default WebChat provider. Live web search, channels, cloud image generation, and third-party APIs need the credentials you choose to configure.

How is this different from a single-agent terminal assistant or a chatbot? Those are one capability. disp8ch is the whole workspace around them: visual workflows, scheduled automations, multi-agent operations, an org/company control plane, a decision council, memory and skills, research, and design — all driven from plain-English WebChat and a browser UI.

Do I still need a separate document chat tab? No. Data Sources manages uploads, crawls, notebooks, notes, outputs, and citations. WebChat is the single ask/synthesis surface, so document questions can become tasks, workflows, council sessions, designs, or organization goals without copying context between tabs.

Can I run more than one organization/company? Yes. One deployment can host multiple organizations with their own agents, goals, budgets, and governance.

Can I bring work from the app I already use? Yes — import compatible skills, workflow JSON, and company/org templates when you want them in the same workspace. See Migration and Imports.

Does it work unattended? Yes — cron schedules, signed webhooks, agent heartbeats/wakeups, and standing goals with a background daemon keep work moving without you in the loop. Risky and external actions stay confirmation-gated.

Is my data private? It is local-first. Your database, memories, documents, and chat history stay on your machine; only the model/tool/channel calls you explicitly configure leave it.

Can I reach it from my phone or messaging apps? Yes — run it on your machine or a server and talk to it from WebChat or connected channels (Telegram, Discord, Slack, WhatsApp, and more) while it works.