Community · 7 tools · free
Try a complete evidence loop on one real task: agree the scope, authorize work, record what happened, and challenge an unsupported claim. No card, trial clock, or paid memory engine.
For people doing real work with coding agents
Pick up where your agent left off.
Keep the evidence, save the next step, and recover your workflow after an interruption. Living Stack adds explicit memory, checkpoints, and signed handoffs to your local agent tools.
One private workspace · up to five installations · no subscription.
Requires Node.js 22+ and an MCP-capable host. Your agent must use the tools; this is not automatic memory.
A real free proof, not the whole product
Community proves the idea. Complete Local runs the work.
Complete Local · 23 tools · $19 once
Adds durable context memory, checkpoint recovery, Ed25519-signed portable traces, release-byte verification, adoption receipts, and the full multi-agent Team surface.
The reason to upgrade
Your task spans more than one sitting. You need the saved context and next step after a restart, a signed record to hand over, or a shared policy across agents. That is the paid workflow.
Buy a workflow, not a tool count
Keep the work moving.
Resume an interrupted build
Store the relevant evidence and a checkpoint before stopping. After a restart, retrieve that same session, load the saved next step, and recheck what is still true.
Includes scoped context and recovery checkpoints. It does not recreate work you never saved.
Hand over a verifiable result
Bind an outcome to the exact target, check the release bytes, and export a signed, metadata-only trace for another person or installation to verify.
A signature proves installation origin and consistency, not that every supplied claim is true.
Coordinate a small agent team
Use one versioned workspace policy across up to five registered installations. Check trace signatures and distinguish configured lifecycle coverage from observed coverage.
The commercial Team layer adds coordination; Community stays a single proof loop.
One complete local runtime
23 tools, bounded. One exact paid bundle.
Before action
Freeze scope, goal, risk policy, retention, and budget. Issue typed authorization receipts before the host acts.
After action
Bind exact evidence to exact targets, reject unsupported completion claims, save recoverable checkpoints, and export Ed25519-signed traces.
Across agents
Coordinate up to five installations under one policy and distinguish installed from active, stale, drifted, or broken lifecycle coverage.
Enforced mode (opt-in)
Hooks that can say no.
Off by default. Turn it on with living-stack-team-verify plan --enforce, then install. It ships in the Team component (0.2.0-beta.2) of the Complete Local 0.3.2-beta.1 bundle.
Completion claims
Claude Code cannot end a turn by claiming the work is done unless a test run the hook itself watched passed after the last edit.
Destructive actions
Destructive commands and writes outside the project folder are blocked unless you authorized that exact action.
Bench result
On 13 recorded hook inputs, it blocked 10 of 10 risky cases and 0 of 3 safe actions. On the same inputs, the best free hook tools we tested blocked at most 3 of the 10 each.
What it does not stop: writes made from inside interpreter one-liners or scripts (python -c, node -e, a script file); tests weakened until they pass trivially; a stop allowed after 3 blocked attempts in a row (recorded as unverified, with a warning); anyone who can use the local installation key re-signing the evidence log. Assertion counting covers node:assert only; other runners get pass/fail checks. Codex support is covered by unit tests only, not by the bench. Like the rest of Living Stack, Enforced mode does not guarantee agent behavior.
Price
$19 once.
No subscription. One private local workspace and up to five registered installations.
Delivery
Exact-hash fulfillment.
The complete archive and license are released only after Stripe verifies payment. Package identity is published before checkout.
Before you buy
Is Complete Local right for you?
I only need occasional claim checks. Do I need to pay?
No. Community can serve that narrower job. Buy Complete Local when interrupted work, portable signed handoffs, release checks, or multi-agent coordination are a recurring need.
What arrives after checkout?
The exact-hash private bundle, installation instructions, and a license for one private workspace with up to five installations. Save your recovery kit before payment. Verify the downloaded archive, install locally, and connect the returned server paths to your host.
Does this replace my agent or run it automatically?
No. Your existing agent and host still do the work and control permissions. Living Stack supplies bounded tools they must invoke. Historical context must be revalidated, and authorization receipts do not force a host to obey them.
How is the free edition different from the paid license?
The current Community download contains seven tools, without memory, checkpoints, signed export, release verification, or Team code. The full single-agent component included with Complete Local retains Apache-2.0 rights; prior full-beta license grants cannot be revoked. The Team component has a separate commercial license. You are buying the assembled bundle and Team capabilities, not exclusive rights over previously open code.
Honest boundary
Proof infrastructure, not a magic model.
Living Stack does not execute host actions, guarantee agent behavior, or turn a configured hook into proven lifecycle coverage. It makes those distinctions measurable.