Python-first action governance for AI automations
DAAI adds policy checks, client approvals, status tracking, and governance receipts to registered Python agent actions - without rebuilding your automation stack.
Explicit SDK-based interception. Deterministic policies. Human approval when it matters.
Proposed action
run_8c4fInput summary
Follow up on overdue invoice INV-1025 for ABC Pty Ltd before a finance workflow sends email.
Policy decision
Policy gatePolicy
Require approval above $500
Decision
Human approval required
Approver
client@example.com
Proposed
Policy evaluated
Approval email sent
Waiting for client
Developers keep execution in their system. DAAI handles proposal, policy, approval, status, and receipts.
Your Python app sends a registered action proposal before the real function runs.
DAAI checks the action, workspace, payload, and deterministic policy rule.
The run becomes executable, pending approval, rejected, or blocked.
Your worker keeps ownership of business execution and external integrations.
The decision, policy snapshot, status, and execution result become audit evidence.
Built for consultants who need client trust, auditability, and safe execution paths.
Define the risky actions your automation is allowed to propose.
Runtime decisions are rule-based, not guessed by an LLM.
IF
amount > $500
THEN
require approval
Send secure approval and rejection links before execution.
Approval request
send_invoice_reminder
Finance agent wants to send a reminder for INV-1025.
Add a gate before risky functions with a few lines of Python.
from daai import DaaiClient
ticket = manager.propose(
action="send_invoice_reminder",
actor="finance_agent",
input=payload,
)
if ticket.executable:
send_invoice_reminder(payload)Every governed decision becomes a receipt.
Execute later only after DAAI says the action is executable.
pending
approval email sent
approved
ready for worker
executed
reported by app
Track what the agent proposed, why it was risky, who approved it, and whether it actually executed.
Workspace
24
action runs
7
pending approval
14
executed
3
blocked
| Action | Actor | Policy | Status | Execution | Time |
|---|---|---|---|---|---|
| send_invoice_reminder | finance_agent | approval > $500 | Pending approval | Not executed | 2m ago |
| update_invoice_status | finance_agent | log only | Allowed | Executed | 11m ago |
| mark_invoice_paid | finance_agent | always approval | Approved | Awaiting report | 18m ago |
| unknown_action | agent | not registered | Blocked | Not executed | 32m ago |
Python SDK
DAAI fits into the safe point before your automation performs a real-world action.
1import os2from daai import DaaiClient3from daai.store import SQLitePendingStore4from daai.runtime import PendingActionManager, DaaiActionRunner5 6client = DaaiClient(7 api_key=os.environ["DAAI_API_KEY"],8 workspace_key=os.environ["DAAI_WORKSPACE_KEY"],9)10 11store = SQLitePendingStore("daai_pending.db")12manager = PendingActionManager(client=client, store=store)13 14ticket = manager.propose(15 action="send_invoice_reminder",16 actor="finance_agent",17 input={18 "invoice_id": "INV-1025",19 "customer_name": "ABC Pty Ltd",20 "amount": 1250,21 },22 reasoning="Invoice is overdue and needs a polite follow-up.",23)24 25if ticket.executable:26 send_invoice_reminder()DAAI separates proposed intent from actual execution, so clients can approve risky actions before anything irreversible happens.
Registered action and payload
Policy produces allowed, pending, or blocked
Secure one-time approval link
Only after executable is true
Decision and execution result recorded
Start with the limits that are enforced by the backend today. Paid tiers will be available after a successful beta test.
Paid tiers will be available after a successful beta test.
Free beta access is intentionally limited during launch to protect system reliability.
$0/month
For proving the full approval and receipt path with real clients.
These limits are enforced in the backend during the free beta.
After beta
For consultants governing a small set of client automations after launch validation.
Paid tiers will be available after a successful beta test.
After beta
Built for small agencies standardizing governance after the beta proves reliability.
Paid tiers will be available after a successful beta test.
Leave an email and a short note. We will follow up when paid tiers are ready after beta.
Start with one client workspace. Register one risky action. See the full approval and receipt flow.