01 · Connect
Work from the tools you already use.
Without it, OrgX becomes one more app your team forgets to update.
WorkingConnect
Claude Code · nowLaunch package session recorded
WritingFor agencies delivering client work with AI agents
Hand client work to agents.Show the client who signed off.
Guardrails before the work. Acceptance after it.
Claude Code, Codex, Cursor, and OpenClaw work inside limits you set. Sends, deploys, and spend always stop for a person, and each deliverable leaves with its evidence and the owner who accepted it, in a proof room with your brand on it.
YouReview essential decisions
Escalateconsequential decisions
Intent & rulesconstraints
ORGXLaunch narrative v5
Accepted- Decisions
- 1
- Artifacts
- 3
- Checks passed
- 2 / 2
Example record
proof/demoAnatomy of an accepted outcomeGateways govern the call.
Gateways govern the call.
OrgX keeps the record of the outcome.
- 01You name the outcomeOne sentence of intent, with the line you will not cross.
- 02It becomes owned lanesEach lane gets an owner in the client it already runs in.
- 03Every client writes one recordSessions, artifacts, and handoffs land in the Work Ledger as they happen.
- 04Judgment pauses only what depends on itA decision waits for you. The lanes that do not need it keep moving.
- 05Checked against your barQuality Studio checks every artifact. An unsupported claim is caught and repaired.
- 06Accepted, with evidenceThe record seals into proof you can hand to a buyer, an auditor, or the next agent.
One record only works when every layer holds.
OrgX connects the layers between assigning work and knowing it worked. Remove one and the record stops being trustworthy.
One outcome, end to endDelegate one outcome
Needs youPlan approval required
Outcome
The launch holds until you approve it. That boundary is the product.Prepare a buyer-ready launch package. Hold the buyer promise and publish for my approval.
ProductPricing
Choose the primary buyer path
PaceReady
DesignLaunch experience
Verify 375, 768, 1024, and 1440 viewports
DanaReady
MarketingLaunch narrative
Finalize launch narrative v5
MarkReady
Start in the agent you already use
Get OrgX working in about two minutes.
Choose a client. Copy one setup, then send your first outcome.
02 · Define the quality bar
Teach OrgX what good means.
Your checks, references, and judgment become the published rule.
Quality StudioEnterprise launch packageBuyer-facing release artifacts · agency leads
Needs youSet your release checks
03 · See what needs you
Only the consequential decision interrupts.
Eligible work keeps moving. Only the dependent work waits for you.
Live workEnterprise launch readinessInitiative → lane → task
WorkingThree work lanes executing
PacePricing
WorkingChoose the primary buyer path
DanaLaunch experience
WorkingVerify 375, 768, 1024, and 1440 viewports
MarkLaunch narrative
WorkingFinalize launch narrative v5
Dana is verifying responsive readiness while the buyer-promise lanes wait for their decision.
Active task · Dana · via CursorVerify responsive launch package
Working375 and 768 complete · 1024 and 1440 in review
- 375
- Passed
- 768
- Passed
- 1024
- Reviewing
- 1440
- Queued
04 · Keep the proof attached
Inspect the work and the proof.
The artifact carries the bar, the repair, and the owner decision.
Proof roomLaunch narrative v5Enterprise launch readiness
ENTERPRISE LAUNCH · BUYER PROMISE
Every release ships audit-clean—automatically.
Automated checks guarantee every customer requirement is met before launch.
Request a walkthroughRead the release notes
OutcomeBuyer-ready launchQuality ruleBuyer-facing launch · v5
Unsupported guarantee“Every” and “automatically” are not supported by the attached release evidence.
05 · Choose the consequence
Start with one initiative. Expand when the stakes do.
Run the first governed loop free. Add capacity yourself, or have OrgX install a client-delivery loop with you.
Software plans
What must the record defend?
$98/month · billed monthly
Keep delegating without reconstructing the work.
What it unlocks
- Unlimited initiatives with durable work records
- Approval-gated delegation across connected tools
Usage and limits
- Unlimited initiatives
- 500 agent credits each month
Managed first loop
Want OrgX to run the first proof loop?
One recurring client-delivery workflow, one buyer-ready proof room, and one durable approval-gated agent loop.
$2,500/monthOne workflow · first proof room within 7 days
Start with one real initiativeDelegate the work. Keep the proof.
Research, releases, and field notes
All posts- · Field noteGuardrails Before the Work, Acceptance After ItHanding real work to agents takes two records: what an agent may do before it acts, and who accepted what it produced afterward. OrgX keeps both on the same initiative.
- · Field note3,328 receipts. One acceptance row.Braintrust and Basis made agent behavior testable. The next primitive makes the judgment cumulative: an open work receipt that binds a behavior verdict to authority, artifacts, cost, and human acceptance.
- · Open standardAgent work now has a receiptAgent Work Receipt v0.1 is an open, account-free contract for portable evidence, deterministic integrity, and independent verification across agent runtimes.
Questions
Before you start
Still deciding? Bring one piece of agent work to a 30-minute review.
Book a proof reviewWhat does OrgX do?
OrgX is proof for AI-delivered work. Claude Code, Codex, Cursor, and OpenClaw write to one shared record, so a decision keeps its provenance (the sessions, the clients, the evidence) no matter which agent picks the work up next. Capture depth still differs by client today.
Can we control what agents are allowed to do?
Yes. Each initiative runs in manual, gated, or autopilot mode, and each connector is set to allow, ask, or deny. Sending, spending, deploying, merging, and deleting always stop for a person in every mode. Checking finished work against a published quality bar is in beta.
Do we have to switch AI tools?
No. OrgX connects through MCP and plugins to Claude Code, Codex, Cursor, ChatGPT, and OpenClaw. Setup differs by client; the record is shared across all of them.
What counts as “accepted” work?
An agent reporting done is a claim, not an approval. An agent can never approve its own work: that floor is enforced in code and fails closed. Work counts as accepted when an owner records the ruling.
Can someone outside OrgX verify the work?
Agent Work Receipt v0.1 is an open, account-free format. A receipt carries its evidence and integrity hash, and it can be checked with the public validator without an OrgX login.
What does it cost to start?
The Free plan covers one active initiative. Paid plans start at $98 a month. If you want us to run it with you first, two scoped episodes on one of your real deliverables cost $250.