Growth agents
buy ads.
“$185 on Instagram” — does it reach the enterprise developers the brief asked for?
Agent spend authority · Shadow Mode is free
Your agents can spend.Neltava decidesif they should.
For teams whose agents buy ads, data, API credits and compute for their customers. Before money moves, Neltava checks every spend against the agent’s rules and the purpose you gave it — then records why.
Observing — every spend goes through, each one is decided and recorded.
illustrative · hover a spend
- decided
- 0
- observed
- $0
- would review
- 0
- would block
- 0
The missing check
Limits check the amount.Networks check the mandate.Neltava checks the purpose.
A $185 Instagram campaign sits inside every budget and limit. It is also no use to an agent hired to find enterprise developers. Only a purpose check notices — and it sends the spend to a person instead of blocking it.
Who it’s for
Agents that spend other people’s money.
Platforms whose agents buy on a customer’s behalf — ads, data, compute, credits. Every spend decided against that customer’s brief, and a record you can show them.
Research agents
buy data.
“$630 to an unknown vendor” — outside the merchants this agent may pay.
Ops agents
buy compute.
“$2,400 on a GPU cluster” — capped at the $800 per-transaction limit.
Builder agents
buy API credits.
“$200 in model credits” — within budget, on purpose. Allowed, and recorded.
in minutes→
How it decides
Inside the moment before money moves.
01 · Ask
Your agent asks first.
One request before it pays: the agent, its task, the steps that led here and the spend. Nothing about your payment rail changes.
growth-agent · run 4812 · step 3
POST /v1/decisions{ "agent": "growth-agent", "task": "Q4 pipeline: enterprise developer leads", "action": { "merchant": "Instagram Ads", "amount_minor": 18500, "description": "Consumer social campaign" }}
02 · Rules
Rules decide what rules can.
Budget, per-transaction limit, merchant, category, velocity — checked the same way every time. Rules can allow, cap or block. Only rules can block.
Authority v7 · deterministic
- Budget$4,815 of $50,000 this month✓
- Per-transaction$185 of $500✓
- Merchantad platforms · allowed✓
- Categoryadvertising · allowed✓
- Velocity3 of 20 this hour✓
rules: pass · the amount is fine
03 · Purpose
Then: does it serve the purpose?
Every agent carries a written purpose. When the fit is in doubt, the spend goes to a person. A model never blocks money on its own.
Purpose · written by you
Acquire qualified enterprise developer leads.
Constraint: no consumer social campaigns
The spend
$185 · Instagram Ads · Consumer social campaign
REVIEW
purpose in doubt
04 · A person
A person makes the call.
Reviewers allow or deny in the console, and label whether the spend served the purpose — the ground truth Neltava is measured against.
Review queue · console
reviewgrowth-agent wants to spend $185 at Instagram Ads.
Within limits. Purpose in doubt: consumer campaign vs. enterprise developer leads.
AllowDeny
label: does not serve the purpose · by a person
05 · Record
And the record holds up.
Each decision keeps its authority version, policy version and a step-by-step trace. Hash-chained per workspace: tamper-evident, verifiable on demand.
#41 · c04a…
#42 · 1b7e…
#43 · 9f2c…e1a0
- decision REVIEW → person DENY · label MISALIGNED
- authority v7 · policy v3 · trace 6 steps · reproducible
- chain verified ✓ · tamper-evident
Shadow, then Enforce
Earn the right to block before blocking anything.
In Shadow Mode Neltava decides every spend and changes nothing — your agents keep running. A week later the Shadow report shows what it would have stopped, what it would have sent to a person, and whether the evidence is strong enough to enforce.
$18,420
autonomous spend observed
$1,630
outside delegated authority
$940
of that, actually spent
illustrative week · “prevented” counts rules only, never a purpose assessment
What happenedamountNeltava, in shadow
- Clearbitresearch-agent · paid ✓$49allow
- Instagram Adsgrowth-agent · paid ✓$185would review
- Unknown vendorresearch-agent · paid ✓$630would block
- GPU clusterops-agent · paid ✓$2,400would cap
- Platform Weeklygrowth-agent · paid ✓$420allow
- DevConf boothgrowth-agent · paid ✓$1,200would review
Ready for Enforce when
- ✓Observed for≥ 7 days
- ✓Agent actions≥ 100
- ✓Purpose coverage≥ 99%
- ✓Human labels≥ 20
- ✓Disagreement with people≤ 10%
Explicit criteria, not a confidence score.
growth-agent
Shadow
A person switches Enforce on — one agent at a time.
Shadow is free.
Enforce when the evidence says so.
Pricing
Shadow
Freeno card
See what your agents would have done.
- Every agent spend decided and recorded
- 30 days of purpose checks from your first decision, up to 5,000
- Shadow report and Enforce readiness criteria
- Human reviews, labels and a verifiable history
Enforce
from $499per month
Let Neltava's decisions take effect.
- Everything in Shadow, without the 30-day limit
- Enforce, switched on per agent
- A review queue for spends that need a person
- Outcomes and signed webhooks
- Set up with the founders
Straight answers.
No. Neltava decides; your agent pays the way it already does. There is no payment migration and no new card.
No. A purpose assessment can only send a spend to a person. Blocking takes a deterministic rule you set — budgets, limits, merchants, categories, velocity — and only in Enforce.
Any. The check happens in your code, right before the payment call — cards, invoices, ad platforms and API credits work the same way.
Your agents keep running and every spend is still decided against your rules and recorded. Purpose checks pause until you move to Enforce.
What your agent sends with each decision, the decision and its trace, and your team's reviews and labels — kept per workspace.