Taskflo is in development and talking with enterprise teams in regulated industries about their first workflows. Product status

taskflo
Talk to us

Explore Taskflo

Talk to us

GOVERNED AI AGENT WORKFLOWS

The trust layer for enterprise AI agents.

Your teams want AI agents doing real work. Your security, risk and compliance teams keep saying no, and they’re right to. Taskflo is being built to give them a defensible way to say yes.

Talk to us about your workflowSee how it works ↓

DESIGNED FOR YOUR CLOUD.
YOUR RULES. YOUR PEOPLE IN CONTROL.

REFERENCE WORKFLOW / 01 Invoice triage
INPUT Invoice received
  1. Invoice triaged
  2. Separate approval above $500
  3. Post to ERP
  1. ExtractInvoice fields
  2. ValidateSchema & vendor
  3. Triage agentClassify exception
  4. CheckpointDurable state
  5. BranchAbove $500?
  6. ApproveSeparate approver
  7. Post to ERPDeclared action
Below threshold: Branch → Post to ERP
Above threshold: Branch → Approve → Post to ERP
BOUNDS BEFORE EXECUTION
Maximum cost per run
$2.50
Reachable systems
3
Irreversible action
1
Human approval
1
Illustrative design values for Taskflo's invoice-triage reference workflow. Not a live run or a customer result.

01 / THE GAP

A working pilot isn’t a permission to deploy.

Cost. Reach. Accountability. Before an agent does real work, someone needs answers they can defend.

02 / THE QUESTIONS THAT MATTER

Three questions between an agent and production.

01

Where does our data go?

The runtime belongs in your cloud account.

Review the boundary02

Who can act on our behalf?

Only declared tools, within limits your teams set.

Inspect the four checks03

Can we prove what happened?

Every action needs a tamper-evident record.

See the evidence model

Score any platform against these questions.

The AI Agent Approval Playbook ↗

03 / HOW IT WORKS

Two halves. The split between them is the whole idea.

Design in Studio. Run in your cloud. The sealed bundle connects them, without handing over your runtime.

WE HOST THE DESIGN PLANE

Taskflo Studio

Design the work.
Declare its authority.
Review its bounds.

workflowpinned componentspolicy packsignature
Sealed bundlePublished to your registry.
Pulled by your engine.
YOUR CLOUD ACCOUNT

Taskflo Execution (TFX)

The engine runs the reviewed version.

Identity & policy
Your keys
Your systems
Audit ledger

Authorise → Comply → Execute → Attest

Studio publishes a signed design bundle. TFX, model access, credentials and runtime systems sit within the customer-controlled boundary. Optional, redacted run metadata is a separate opt-in channel.

AT EVERY ACTION

Four checks. In order. Every time.

  1. 01

    Authorise

    Is this agent allowed to act?Cedar · access policy
  2. 02

    Comply

    Is it within your rules?OPA / Rego · policy packs
  3. 03

    Execute

    Act within the agreed bounds.Scoped credentials · hard caps
  4. 04

    Attest

    Record it before returning it.Hash-chained record

Denied at any check? Stop the action. Name the rule. Record the decision.

Explore the platform ↗Five principles. Five mechanisms. ↗

04 / PROOF BEFORE IT RUNS

The page your reviewer should see before go-live.

What it can spend. What it can reach. Which actions need a person. These are the bounds Taskflo is designed to calculate for each version.

For financeA declared spend ceiling.

For securityA named reach list.

For the reviewerBounds attached to the version being approved.

How the bounds are calculated ↗
REFERENCE WORKFLOW

Invoice triage

DECLARED BOUNDS / BEFORE GO-LIVE
$2.50
Maximum cost per runA declared ceiling, calculated from the workflow design.
3
Reachable systemsThe systems this workflow is designed to contact.
1
Irreversible actionNamed before promotion, with recovery requirements reviewed.
1
Human approvalA separate approver for invoices above $500.
Illustrative design values for Taskflo's invoice-triage reference workflow. Not a live run or a customer result.

06 / THE REGULATORY CONTEXT

The EU AI Act turns good practice into law.

Technical controls need to produce evidence. Your use case, role and transition provisions determine which obligations apply.

  1. 2 August 2025General-purpose AI rules
  2. 2 August 2026Broad application, including transparency duties
  3. 2 December 2027Annex III high-risk obligations
  4. 2 August 2028High-risk AI integrated into Annex I regulated products

€35m / 7%Prohibited practices

€15m / 3%Specified operator obligations

€7.5m / 1%Incorrect information to authorities

Application depends on use, role and transitional provisions. Fine ceilings depend on the infringement and undertaking; special rules apply to SMEs. Regulation (EU) 2026/1744 · AI Act, Article 99 · European Commission timeline.

ONE WORKFLOW. A CLEAR NEXT STEP.

Everyone is building agents. We’re building the reason a bank is allowed to run one.

Tell us about the process, the systems and the people. Let’s work through what it would take to run it.

Talk to us about your workflow