DEV.co
Open-Source DevOps · BradGroux

veritas-kanban

Veritas Kanban is a local-first task management and Kanban board built in TypeScript that optionally orchestrates AI agents. It starts as a standalone board and layers on CLI, MCP server, webhooks, and governance features only when needed.

Source: GitHub — github.com/BradGroux/veritas-kanban
751
GitHub stars
93
Forks
TypeScript
Primary language
MIT
License (OSI-approved)

Key facts

Objective fields from the source. Values we can't verify are shown as “Unknown” rather than guessed.

FieldValue
RepositoryBradGroux/veritas-kanban
OwnerBradGroux
Primary languageTypeScript
LicenseMIT — OSI-approved
Stars751
Forks93
Open issues4
Latest releasev5.2.1 (2026-06-29)
Last updated2026-07-06
Sourcehttps://github.com/BradGroux/veritas-kanban

What veritas-kanban is

TypeScript/React-based Kanban board with a Node.js backend, optional MCP server (36 tools), CLI, Squad Chat webhooks, and agent provider routing (Codex, Ollama, LM Studio). Data is stored locally-first; external integrations are opt-in layers.

Quickstart

Get the veritas-kanban source

Clone the repository and explore it locally.

terminalbash
git clone https://github.com/BradGroux/veritas-kanban.gitcd veritas-kanban# follow the project's README for install & configuration

Need it deployed, integrated, or customized instead? DEV.co ships production installs.

Best use cases

AI Agent Task Orchestration

Coordinate multiple AI agents (Claude, GPT, Ollama) on shared tasks with audit trails, approval gates, and role-based governance—useful for multi-model code review and agent handoffs.

Local-First Team Task Management

Run a Kanban board entirely locally (no cloud vendor lock-in) for small teams or individual developers who want visual task tracking with optional CLI and webhook integration.

Self-Hosted DevOps & Agent Workflow

Deploy on your own infrastructure for regulated environments, customize workflows via MCP tools and policy engines, and maintain full audit trails for compliance-sensitive work.

Implementation considerations

  • Start with board-only setup (pnpm install, .env, localhost:3000) before adding CLI, MCP, or agent layers; docs emphasize staged adoption to avoid configuration bloat.
  • Agent provider routing requires explicit setup per provider (Codex SDK, Ollama endpoints, LM Studio config); no auto-detection—read Agent Providers and Codex Integration docs.
  • Local task storage uses .md files in tasks/ directory; backup strategy and conflict-merge behavior on multi-client writes not clearly specified.
  • v5 desktop app (Homebrew tap) exists but desktop architecture ADR and mobile offline ADR indicate active post-GA development; confirm desktop stability for your use case.
  • MCP server (36 tools) requires write-access security review; docs mention security model but no pen-test results or CVE history provided.

When to avoid it — and what to weigh

  • Highly Distributed, Large-Scale Teams — No mention of conflict resolution, real-time sync, or multi-region support. Local-first design and documented setup paths suggest primary target is small/local groups.
  • Strict Enterprise RBAC/SSO Requirements Out-of-Box — v5 identity model exists but is still in development (see v5 GA Checklist). Fine-grained permission matrices may require custom configuration or post-GA work.
  • Zero DevOps Appetite — Self-hosting, environment setup, MCP server config, and agent provider routing all require technical effort. Not a plug-and-play SaaS for non-technical users.
  • Vendor Lock-In Avoidance with Full Feature Parity — Optional layers (OpenClaw, workflows, governance) tie task data to specific agent APIs and custom schemas. Full export/portability story not clearly documented.

License & commercial use

MIT License: permissive, allows commercial use, modification, distribution, and private use with attribution. No patent grant, no warranty, no liability.

MIT permits commercial use without restriction. However, no warranty is provided—verify security posture, performance guarantees, and support terms separately. Project is community-driven (single creator, 751 stars); no SLA or commercial support model documented.

DEV.co evaluation signals

Editorial assessment — not user reviews. Directional, with an explicit confidence level.

SignalAssessment
MaintenanceActive
DocumentationStrong
License clarityClear
Deployment complexityModerate
DEV.co fitGood
Assessment confidenceHigh
Security considerations

Local-first storage reduces external exposure; .env pattern for admin keys follows best practice. MCP server exposes 36 tools with write access—authorization model and input validation not detailed in excerpt. No mention of audit logging, encryption-at-rest, or threat model. Self-hosted deployment requires hardening (docs point to Self_Host.md). Desktop app and native mobile (post-GA) need security review before sensitive use. No CVE history or security contact data provided.

Alternatives to consider

Jira / Linear

Cloud-hosted, enterprise-grade RBAC, SSO, and audit. Trade-off: vendor lock-in, no local-first option, higher cost. Better if team >10 and needs out-of-box compliance.

Trello + Zapier / Make

Simple visual boards, broad integrations, no setup required. Trade-off: limited agent orchestration, data lives in Atlassian cloud, no local-first. Better for non-technical teams or light workflows.

Notion + Custom Automations

Flexible, familiar, good API. Trade-off: no native agent orchestration, Notion lock-in, slower performance at scale. Better if team already in Notion ecosystem and agents are secondary.

Software development agency

Build on veritas-kanban with DEV.co software developers

Clone the repo, run pnpm dev, and start with the board-only setup. Add agent layers when you need them—docs walk you through each step.

Talk to DEV.co

Related open-source tools

Surfaced by semantic similarity across the DEV.co open-source index.

Related on DEV.co

Explore the category and the services that help you build with it.

veritas-kanban FAQ

Can I use Veritas Kanban without running a backend server?
Not fully. The board UI and CLI require a Node.js backend (localhost:3001). Board-only setup is minimal (pnpm dev), but you cannot run UI-only in a browser without a server process.
What happens to my tasks if I add AI agent integrations later?
Tasks are stored locally in tasks/ .md files. Adding MCP, Squad Chat, or agent layers is additive—existing board state persists. However, schemas for workflows and governance are not documented as backward-compatible; verify upgrade path for your version.
Is there a hosted version or do I have to self-host?
No hosted SaaS documented in the excerpt. Post-GA Cloud Sync and Hosted SaaS ADR is in the roadmap but not shipped. Today: local dev, desktop app (macOS via Homebrew), or self-host on your infrastructure.
Can multiple team members work on the same board in real-time?
Unknown. Local-first architecture and .md file storage suggest single-writer or eventual-consistency model, but multi-client sync behavior, conflict resolution, and real-time collaboration are not clearly documented in the excerpt. Requires review of sync internals or testing.

Software development & web development with DEV.co

Need help beyond evaluating veritas-kanban? DEV.co is a software development agency offering software development services and web development for teams of every size. Our software developers and web developers build custom software, web applications, APIs, and open-source devops integrations — and maintain them long-term.

Ready to orchestrate your AI workflow?

Clone the repo, run pnpm dev, and start with the board-only setup. Add agent layers when you need them—docs walk you through each step.