Agent report card
Can an agent use Neon API through Project and database lifecycle?
Five-stage assessment
Where agent autonomy holds and where it breaks
- EvaluateLimited
- Sign upLimited
- PayReady
- ProvisionLimited
- OperateLimited
Discovery works but is incomplete, context-dependent, split, or unstable.
The page states that the Neon API reference has moved to https://neon.com/docs/reference/api and that api-docs.neon.tech is deprecated and will be retired soon, though the Neon API itself remains fully supported and unchanged.
Applicable terms are stable, readable, retrievable, and materially complete.
Eligibility conditions and required inputs are explicit and decidable.
Price or no-charge status, variables, currency, and material conditions are explicit.
Verified structured artifacts accelerate discovery of the evaluation surface.
Evidence and actions6 signals
Can an agent decide every material eligibility condition before commitment?
supported: freshEligibility conditions and required inputs are explicit and decidable.
The Neon terms of service establish that Databricks and Customer govern the use of Platform Services. Entering into the Schedule on behalf of a company requires representing and warranting authority to bind that entity, otherwise 'Customer' refers to the individual.
Can an agent estimate cost or confirm no-charge status before commitment?
supported: freshPrice or no-charge status, variables, currency, and material conditions are explicit.
Neon explicitly details its pricing rates, variables, and currency in USD. For example, Launch plan compute is $0.106 per CU-hour, Scale plan compute is $0.222 per CU-hour, and Free plan provides 100 CU-hours of database compute per project per month scaling to zero when idle.
Can an agent find the exact service and its stable entrypoints?
supported: freshDiscovery works but is incomplete, context-dependent, split, or unstable.
The page states that the Neon API reference has moved to https://neon.com/docs/reference/api and that api-docs.neon.tech is deprecated and will be retired soon, though the Neon API itself remains fully supported and unchanged.
Are verified structured discovery artifacts available?
supported: freshVerified structured artifacts accelerate discovery of the evaluation surface.
The documentation explicitly references structured discovery artifacts, noting that an index of all pages formatted in Markdown and endpoints in OpenAPI is available at https://api-docs.neon.tech/llms.txt.
Can an agent retrieve and understand the applicable commitment terms?
supported: freshApplicable terms are stable, readable, retrievable, and materially complete.
Applicable terms are publicly accessible and retrievable via the Product Specific Schedule (Neon), which governs Customer's use of Databricks' Neon product family and incorporates the Databricks Master Cloud Services Agreement.
Does the vendor deliberately describe access for web agents?
supported: freshCurrent admissible evidence does not resolve this finding.
The retained evidence does not provide explicit declarations or policy rules regarding web agent or web crawler access permissions such as robots.txt directives.
What would improve this stage
- Publish and link one stable canonical service and access entrypoint.
Machine credentials exist but require manual copying, broad authority, or weak lifecycle support.
Neon supports Personal, Organization, and Project-scoped API keys with specific roles and revocation support, but creating the first API key requires manual authentication in the console.
A stable canonical route begins the required access bootstrap.
Evidence and actions5 signals
Can an agent operate the access controls and safe handoffs deterministically?
supported: freshCurrent admissible evidence does not resolve this finding.
The signup endpoint returned a 403 Cloudflare security verification page, leaving the operability of access controls unresolved.
Is there a stable route to begin obtaining service access?
supported: freshA stable canonical route begins the required access bootstrap.
Neon documentation confirms stable access entry points including Sign up, Log in, and Get started buttons.
Can an agent obtain access without an unsupported CAPTCHA boundary?
supported: freshCurrent admissible evidence does not resolve this finding.
The signup page returned a Cloudflare bot security verification page, and documentation does not explicitly declare a verified-agent alternative or CAPTCHA policy.
Can an agent obtain scoped, revocable authority for this service?
supported: freshMachine credentials exist but require manual copying, broad authority, or weak lifecycle support.
Neon supports Personal, Organization, and Project-scoped API keys with specific roles and revocation support, but creating the first API key requires manual authentication in the console.
Can required phone verification be completed through a supported boundary?
supported: freshCurrent admissible evidence does not resolve this finding.
The signup page returned a 403 security verification page, and documentation describes account authentication and 2FA options without explicitly addressing phone verification.
What would improve this stage
- Provide scoped delegated authorization with consent, revocation, and deterministic resumption.
Charge or no-charge status, currency, recurrence, and material conditions are disclosed.
Neon discloses its plan commitment details, including a $0/month Free tier and pay-for-what-you-use pricing for Launch ($0.106 per CU-hour, $0.20 per GB-month storage) and Scale ($0.222 per CU-hour) plans.
A documented direct purchase path reaches paid access without a vendor decision point.
Evidence and actions4 signals
Can an agent construct checkout, hand off approval safely, and resume?
supported: freshCurrent admissible evidence does not resolve this finding.
The retained evidence describes Neon's billing summary and invoice views in the console, but it does not disclose a deterministic API or web workflow for an agent to construct checkout, hand off approval safely, and resume.
Is the exact commercial commitment disclosed before authorization?
supported: freshCharge or no-charge status, currency, recurrence, and material conditions are disclosed.
Neon discloses its plan commitment details, including a $0/month Free tier and pay-for-what-you-use pricing for Launch ($0.106 per CU-hour, $0.20 per GB-month storage) and Scale ($0.222 per CU-hour) plans.
Can payment be authorized within scoped agent or explicit human authority?
supported: freshCurrent admissible evidence does not resolve this finding.
The documentation covers plan prices and paying invoices manually via the console, but it does not document a payment rail supporting scoped agent delegation or explicit human-confirmed authorization with receipts.
Is a direct self-service path to paid access documented?
supported: freshA documented direct purchase path reaches paid access without a vendor decision point.
Neon documents a direct self-service path to paid access where users can upgrade from the Free plan to Launch or Scale pay-as-you-go plans without requiring a vendor decision point or sales interaction.
Delivery needs additional human transfer or has weak scope or lifecycle semantics.
When creating API keys from the Neon Console, the secret token will be displayed only once and cannot be retrieved later, requiring manual copying.
Completion is observable, but an asynchronous path omits a material terminal, bound, or reconciliation property.
Provisioning is machine-triggerable or follows deterministically from an allowed handoff.
Evidence and actions3 signals
Can usable access material be delivered securely to an authorized agent?
supported: freshDelivery needs additional human transfer or has weak scope or lifecycle semantics.
When creating API keys from the Neon Console, the secret token will be displayed only once and cannot be retrieved later, requiring manual copying.
Can an agent determine successful provisioning completion and reconcile asynchronous failure?
supported: freshCompletion is observable, but an asynchronous path omits a material terminal, bound, or reconciliation property.
Create project returns 201 with a connection URI, database, password and role, but the same response documentation instructs callers to wait until the project's operations finish before connecting; the asynchronous completion bound is not established.
Can provisioning be initiated within supported agent authority?
supported: freshProvisioning is machine-triggerable or follows deterministically from an allowed handoff.
Creating a project can be initiated directly via an HTTP POST request to the Neon API endpoint (/projects).
What would improve this stage
- Deliver scoped access material through a secure documented machine-usable flow.
- Expose bounded machine-readable terminal status and reconciliation semantics.
Safe failure handling is documented only partially.
The Create project endpoint documentation details response status codes, including safe retries for GET, HEAD, OPTIONS, 503 Service Unavailable, and 423 Locked responses, as well as warning that non-idempotent methods (POST, PATCH, DELETE, PUT) are generally not safe to retry. Safe failure semantics are only partially documented across all operations.
Only part of the credential lifecycle is agent-operable.
Every essential target has a stable documented agent-usable interface alternative.
Request-time authentication is documented and usable under scoped agent authority.
Essential operations have stable readable request, response, and effect semantics.
Evidence and actions6 signals
Is an agent-native protocol interface verified against the essential targets?
supported: freshCurrent admissible evidence does not resolve this finding.
The retained excerpts describe REST or RPC interfaces, SDKs and CLIs for Neon API; they contain no first-party statement that an agent-native protocol interface is absent, and several navigation excerpts mention MCP resources without establishing coverage of the assessed targets.
- project-api
Can an agent manage credential expiry, rotation, revocation, compromise, and recovery?
supported: freshOnly part of the credential lifecycle is agent-operable.
API keys can be listed, created, and manually rotated or immediately revoked via the Console, CLI, or API, including when a key is suspected of being compromised. However, API keys do not expire on a schedule and expiration handling is not provided.
Can an agent handle applicable failure modes safely?
supported: freshSafe failure handling is documented only partially.
The Create project endpoint documentation details response status codes, including safe retries for GET, HEAD, OPTIONS, 503 Service Unavailable, and 423 Locked responses, as well as warning that non-idempotent methods (POST, PATCH, DELETE, PUT) are generally not safe to retry. Safe failure semantics are only partially documented across all operations.
Can an agent authenticate service operations with scoped authority?
supported: freshRequest-time authentication is documented and usable under scoped agent authority.
Neon API operations authenticate using HTTP request headers containing a Bearer token with a 64-bit API key (Personal, Organization, or Project-scoped).
Are essential operation inputs, outputs, and effects stable and decidable?
supported: freshEssential operations have stable readable request, response, and effect semantics.
Essential operations across the API reference, branch operations, project creation, and project operations document request parameters, HTTP methods, endpoints, JSON schemas, response structures, and operational effects.
- project-api
Can an agent perform every essential assessment target through a usable interface?
supported: freshEvery essential target has a stable documented agent-usable interface alternative.
All essential target operations—including getting started reference, branch operations, project creation, and project management—are operable through documented HTTP API endpoints or CLI tools.
- project-api
What would improve this stage
- Document applicable structured errors, retry, cancellation, and reconciliation semantics.
- Provide scoped expiry, rotation, revocation, compromise, and recovery operations.
What works
- Pay: Charge or no-charge status, currency, recurrence, and material conditions are disclosed.
What blocks agents
No stage is established as blocked.
Where autonomy is limited
- Evaluate: Discovery works but is incomplete, context-dependent, split, or unstable.
- Sign up: Machine credentials exist but require manual copying, broad authority, or weak lifecycle support.
- Provision: Delivery needs additional human transfer or has weak scope or lifecycle semantics.
- Operate: Safe failure handling is documented only partially.
What this grade establishes
14 of 14 graded signals have current supporting evidence. 0 of 5 additional barrier checks are verified.Unknown checks are not a pass or a confirmed blocker. This report is not proof of an end-to-end purchase or service session.
How the grade is derived
A through C grades count Limited stages; D and F reflect Blocked stages by lifecycle severity.
Coverage is complete only when every graded signal has supported, fresh, non-conflicting evidence. Barrier and informational signals do not change coverage; only fresh, supported barrier values that passed their admission evidence rule participate in stage outcomes.
Each stage takes its worst graded signal or verified barrier. A fresh, supported mandatory barrier that passed its admission evidence rule can block that stage; incomplete or stale barrier evidence remains context only. The overall grade is derived from the five stage states.
Startup offers
Neon on the startup-offers board
- Up to $1,000 in Neon creditsNeon Startup Program - Self-Funded
- Up to $200K in creditsNeon Startup Program - Venture-Backed
- Up to $200,000 in Neon and Databricks credits valid for 12 months.Databricks Startup Program - Venture-Backed
- Up to $1,000 in Neon credits valid for 12 months.Databricks Startup Program - Self-Funded
Distribution
The grade’s badge
One stable image resolving the record’s current live facts, linking to the canonical page. Embed it once; it keeps itself current. Placement is never evidence and never affects rank.
[](https://sourcey.com/log/neon/agent-readiness/neon-api/project-database-lifecycle)Improve the report
Correct it, rerun it, or improve the funnel.
- Profile
- arp_01kzf0m4xc6w3hq8r2f9bj5nta
- Projection
- sha256:93204be9d7e09cf2622297df3db47e306e9b74bb18f77cfb6599eb52f41c738b
- Policy
- service-use-2026-09-07-blocking-barriers-r12 · sha256:fc9e8f046d86fcf19ab2f1bc547219e503cdc11ffa135d020db7285965cd7d1d
Every request carries these exact identifiers.How report cards are assessed →View the exact declaration