A1 ratingsOpenAIAgent Builder (AgentKit)

Agent Builder (AgentKit)

OpenAI · platform · Any industry
A
AGrade
80Score
±1395% range
≈#7Rank of 1134
0%From fleets
FR-CAT-00025Catalog no.

The six pillars

Each bar is the score; the bracket is its 95% range.

Can you see what it costs?

Vendor cost API (OpenAI Costs API GET /v1/organization/costs) · prices published
75±38
Desk research 100 · counts for 25% · A cost API you can call scores highest; an admin console or cloud billing export next; a card statement or emailed invoice is all some products leave you. Published prices add to it.

Can you see how much it's used?

OpenAI Usage API GET /v1/organization/usage/* (group by project_id, api_key_id, user_id, model)
90±17
Desk research 100 · counts for 15% · A usage API (tokens, seats, credits, conversations) a buyer can read.

Can you tie cost to one agent?

By project
66±17
Desk research 70 · counts for 15% · Exact (one key or project per agent) beats by-project, by-workspace or by-user.

Can an agent on it report out?

Remote MCP connector · Agent Builder workflows can call remote MCP servers and HTTP tools; give each agent its own project or API key for clean attribution.
90±17
Desk research 100 · counts for 15% · A remote MCP connector scores highest, then a local MCP server, an SDK, an API or a workflow step; copy-and-paste lowest. Any open standard counts, not just FleetRegistry.

Can you control spend and access?

Spend cap: Hard cap · Per-agent keys: Scoped · Audit log: By API · Revoke: By API · SSO: SSO + SCIM
75±38
Desk research 100 · counts for 20% · Hard spend caps (30%), scoped per-agent keys (25%), an audit log you can read by API (20%), revoking by API (15%), SSO with SCIM (10%).

Does it work on small plans?

Small plans get the data
90±17
Desk research 100 · counts for 10% · Whether the plan a person or small business buys gets the cost and usage data at all.

Where the evidence comes from

Independent fleets0%Desk research67%Neutral starting point34%

Once three independent fleets report on a pillar, desk research counts for at most a third of it. FleetRegistry's own fleets never count.

Controls on record

Spend capHard cap
Per-agent keysScoped
Audit logBy API
RevokeBy API
SSOSSO + SCIM
PricesPublished

Price changes: Services Agreement: pricing-page changes take effect 14 days after posting; material agreement changes get at least 30 days' notice.

API platform: project/org spend limits can be set as hard limits that return 429 (enforcement can lag slightly); project service accounts with restricted key permissions and per-project model/rate limits; Admin API covers audit logs and deleting keys. SCIM only for API orgs on custom/unlimited billing.

History

11-12 A 8011-13 A 8011-14 A 8011-15 A 8011-16 A 8011-17 A 8011-18 A 8011-19 A 80
First rated 2026-10-02 at 86 (A). Now 80. Published grades are sealed daily into the transparency log (N4.4). Grade held since 2026-10-02.

Every fact, with its rung and source

Each fact stands on its own source and its own rung of the evidence ladder (Rule N2.1). Verified Observed Declared Inferred. The weakest fact in a pillar sets its weight.

FactValueRungCheckedSources
How a buyer sees cost
Cost
Vendor cost API Declared
high confidence
2026-10-15
1 earlier
[1][2] Correct
Prices published
Cost
Yes Inferred 2026-10-15
2 earlier
[1][2][3][4] Correct
Usage API
Usage
OpenAI Usage API GET /v1/organization/usage/* (group by project_id, api_key_id, user_id, model) Declared
high confidence
2026-10-15
1 earlier
[1][2] Correct
Cost per agent
Per agent
By project Declared
high confidence
2026-10-15
1 earlier
[1][2] Correct
How an agent reports out
Reports out
Remote MCP connector Declared
high confidence
2026-10-15
1 earlier
[1][2] Correct
Per-agent keys
Controls
Scoped Inferred 2026-10-15
2 earlier
[1][2][3][4] Correct
Audit log
Controls
By API Inferred 2026-10-15
2 earlier
[1][2][3][4] Correct
Revoke access
Controls
By API Inferred 2026-10-15
2 earlier
[1][2][3][4] Correct
Spend cap
Controls
Hard cap Inferred 2026-10-15
2 earlier
[1][2][3][4] Correct
SSO
Controls
SSO + SCIM Inferred 2026-10-15
2 earlier
[1][2][3][4] Correct
Small plans get the data
Small plans
Yes Declared
high confidence
2026-10-15
1 earlier
[1][2] Correct

Prices · No separate fee; billed at standard OpenAI API model/tool rates

Notes · AgentKit tools are included with standard API model pricing, so costs land in the org Costs/Usage APIs (admin key). OpenAI is winding down Agent Builder (and Evals); unavailable from 2026-11-30.

Scored with the rubric for Platforms and APIs (Rule N7.1). Each pillar blends a neutral 50, desk facts weighted by rung, and independent fleets. A1 is 85+ with fleet proof, A 70, B 55, C 40, D 25, E below. Part N of the Rules

Run agents on Agent Builder (AgentKit)?

Give each one a permanent number, a named person who answers for it, and a record of what it can touch and what it costs. Free.