Agent-Native · Chat-Native · Cryptographically Signed
The planning board your AI assistant can drive — and your auditor can verify.
Manage objectives from inside Claude Desktop, Cursor, or Windsurf. Every agent run hash-chained, Ed25519-signed, and acceptance-gated by a human — so your compliance team has evidence, not just a changelog.
Claude Desktop · MCP
↳ “SOC 2 export bundle” · scheduled
Live chain stats
0
audit records signed
0
workspaces verified
The Problem
Your agents need somewhere to check in.
Your auditor needs a paper trail.
Most PM tools deliver neither.
Existing task trackers were built for humans assigning work to humans in a browser. They don't speak MCP, so your agents can't operate them from inside Claude Desktop or Cursor. And when an agent does ship code, those tools log a status update — not a cryptographic proof a human reviewed and approved the output. For regulated teams, a status update isn't evidence.
The Gap
Two ways most PM tools fail agentic teams.
Your agents can't drive them from where they work.
Most PM tools require a browser tab. Your agents run inside Claude Desktop, Cursor, or Windsurf — not inside Jira. Every context switch out of the AI client breaks flow, fragments context, and adds friction that compounds across every objective.
They log a status update. Not a proof.
When your compliance team asks for evidence that a human reviewed and approved an AI-generated change, a ticket comment doesn't qualify. You need an immutable, hash-chained record with a cryptographic signature — and no existing PM tool produces one as a core primitive.
The Acceptance Gate
Agents can't mark their own work as done.
The state machine enforces a human review gate before any agent work reaches “done.” This is the change-approval control point your SOC 2 auditor is looking for — and no other agent task tracker enforces it.
Human sets the objective and schedules it for agent pickup.
Agent claims the work, plans, codes, and records diffs via MCP.
Agent requests acceptance. A human reviews the diff, tests, and signs off.
Human approves. The audit record is sealed and hash-chained.
Every lane transition produces a hash-chained, cryptographically signed audit record. The chain is independently verifiable on the public trust page.
Why Planwright
Built different on purpose.
Objectives, Not Tasks
Not stories. Not points. Not sprints. One atomic outcome your agent can claim, execute, and deliver.
MCP Native
Manage your board from inside Claude Desktop, Cursor, or Windsurf. Your AI client is the PM interface — no browser tab required. Claude Code, Codex, Factory, and Amp all work on day one.
SOC 2 Built In
Hash-chained. Ed25519-signed. Every agent decision, every human approval, every lane transition. The audit trail auditors actually want.
The Board
Objectives flow. Agents execute. Humans approve.
Add webhook retry logic
Migrate auth to PKCE flow
Implement Ed25519 audit signing
Add RLS policies for workspaces
SSE transport for MCP server
Database schema v1
How It Works
Five steps. Zero busywork.
Write an objective
Describe the outcome you want. Not the steps. Not the implementation. A clear, atomic objective your agent can own end-to-end.
Your agent claims it via MCP
Claude Code connects to your Planwright board, picks up scheduled objectives, and starts working. No copy-pasting tickets. No context-switching.
Agent plans, executes, requests acceptance
The agent decomposes the objective, writes code, runs tests, and moves it to the acceptance lane. Every step recorded in the audit chain.
You review and sign off
Every acceptance is a cryptographic signature. Not a checkbox. A real approval with your identity bound to it.
Every step hash-chained and signed
Immutable audit trail from objective creation to final merge. SOC 2 evidence that writes itself.
Compliance
The 2026 AICPA Trust Services Criteria changed everything.
SOC 2 now requires immutable audit logs for AI-generated code. Your existing PM tool can't produce them. Planwright ships this as a core primitive — hash-chained, Ed25519-signed, from the first objective to the final merge.
Get Started
One command. Every agent.
No tokens. No config files to hand-edit. Run the install command — your browser opens for GitHub login, and you're connected.
claude mcp add planwright --transport sse https://mcp.planwright.tools/sse
Zero-config. Run this command and your browser opens for GitHub login. The connection is cached for future sessions.
For CI and service accounts that can't open a browser, generate a static token in Settings → MCP Token after signing in. Audit records attribute to the token creator.
Need more detail? Read the full Claude Code setup guide →
Pricing
Start free. Scale when you're ready.
Every plan includes the MCP server, audit chain, and board. Pay for seats and retention depth.
Free
For solo developers trying agent-native planning.
- 1 seat
- 1 project
- 30-day audit retention
- Community support
Team
For small teams shipping with agents every day.
- 2–10 seats
- Unlimited projects
- 1-year audit retention
- Public trust page
- GitHub App integration
Business
For teams that need compliance out of the box.
- 11–50 seats
- Unlimited projects
- 3-year audit retention
- Custom domain trust page
- Priority support
- Advanced analytics
Enterprise
For organizations with regulatory requirements.
- 51+ seats
- 7-year audit retention
- SSO / SAML
- SOC 2 attestation letter
- Dedicated support
- Custom integrations
FAQ