Auth0 for AI Agents (Token Vault) by Auth0 by Okta

# Fix list: Auth0 for AI Agents (Token Vault)

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/auth0-ai-agents, the October 2026 research run, assessed 1 October 2026. Grade BB, 71.5 out of 100.

This is everything the published grade says the listing lacks, the biggest possible gain to the total first. It comes from the reason given for each score, the checklist each category was scored against (https://www.anchorterminal.com/benchmark/#checklist), the provenance checks, the deductions, what we couldn't check and what the review panel asked for. A fix counts at the next check, once it's public.

For a coding agent working on Auth0 for AI Agents (Token Vault): work through the items below in the product, its docs and its public pages. Each category gives the reason for its score, with the points each checklist item earned, and the checklist itself, so the gap is the items that earned less than their points. Change the product, not the wording, and keep a note of what you changed and where it's published.

## 1. Payments & pricing, 30 out of 100, up to 8.8 more on the total

Why it scored 30: No x402, MPP or L402 (0). Plan prices are public (Essentials from $35 a month, Professional from $240 for 500 monthly active users), and the agent add-on rule (50 per cent on top of the base) is in Auth0's docs rather than on the pricing page (10). Free plan up to 25,000 monthly active users with no card (20). A person signs up in a browser and creates a tenant (0).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-payments):

The published rubric, also on the [x402 page](https://www.anchorterminal.com/x402/).

- 40, a machine payment protocol (x402, MPP or L402) on the tool's own endpoints. 10 to 30 when it covers only some endpoints or only goes through a third party, and the note says which.
- 20, per-call or per-unit pricing published without a login. 10 for public plan-only pricing, 0 for "contact sales" or prices behind a login.
- 20, a free tier or trial that doesn't need a card.
- 20, autonomous onboarding, meaning an agent can get access without a person signing up in a browser (keyless use, x402, a programmatic key API).

Payment platforms and agent wallets rarely charge for their own API over a machine protocol, so the first line has steps for them, and the highest one that applies counts. 40 when x402, MPP or L402 runs on all their own endpoints, 30 when it runs on part of their own API, 25 when their merchants can accept one, 20 for running a facilitator, 15 for paying as a buyer, and 0 when the only protocol is their own. Merchant acceptance sits above a facilitator because the platform's own customers can charge agents through it, while a facilitator settles for sellers who wire up the protocol themselves. The counter-argument (a facilitator does more for the protocol as a whole) has a point. Each note says which step applied.

Open-source software you run yourself is scored on its hosted or paid option if it has one. A free, self-hosted package with nothing to buy gets 20, 20 and 20 for the last three lines, and 0 to 40 for the first only if it ships a payment protocol.

## 2. Reliability, 75 out of 100, up to 5 more on the total

Why it scored 75: Auth0's own status page at status.auth0.com with per-region history (20). The history view didn't render for us, so we score it as unreadable, and third-party trackers list several regional incidents in the window, among them a sign-in outage on 24 July 2026 and multi-region errors on 6 August and 1 September (5). Rate limits are published per endpoint and per plan in the rate limit configuration pages (15). 429 with X-RateLimit-Limit, X-RateLimit-Remaining and X-RateLimit-Reset headers and back-off guidance are documented in Auth0's docs repository (15). 99.99 per cent SLA on Enterprise, listed on the pricing page (10). Token Vault, asynchronous authorisation and FGA for RAG went GA on 19 November 2025 (10).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-reliability):

Hosted APIs, MCP servers, models and platforms.

- 20, a public status page with component history (Statuspage, Instatus, BetterStack or the vendor's own).
- 0 to 30, the incident record for the last 90 days on that page. 30 for a clean record or trivial incidents only, 20 for minor incidents only, 10 for one major outage (an hour or more of a core API down, or errors across the board), 0 for several. 5 when there's no history we could read, and the note says so.
- 15, rate limits documented with numbers.
- 15, documented 429 or overload handling (Retry-After, backoff guidance), and idempotency keys or safe-retry guidance where writes are involved.
- 10, an SLA published for any paid tier.
- 10, the surface agents use is generally available, not beta or preview.

Local packages, SDKs, frameworks and stdio MCP servers.

- 20, installs from an official package with supported runtimes stated.
- 25, a public CI and test suite, passing on the default branch.
- 0 to 25, open crash or regression issues relative to activity (25 for few and handled, 0 for many, old and unanswered).
- 15, semver discipline and breaking changes called out in a changelog.
- 15, version 1.0 or later, or declared stable.

Protocols are read from their reference implementations, the public facilitators or servers, spec stability and test vectors.

## 3. Agent ergonomics, 71 out of 100, up to 4.7 more on the total

Why it scored 71: The exchange returns one provider token and its expiry, so there's nothing to size (20). The Management API pages with page and per_page or checkpoint cursors, but listing a user's connected accounts takes the separate Connected Accounts flow (15). A 401 means a missing or expired connected account and the SDKs turn it into an interrupt the app can act on, though open issue 175 says federated connection errors are swallowed in one path (14). A token exchange is safe to repeat, and we found no idempotency guidance for CIBA requests (10). Six required parameters on the exchange call, and official SDKs in JavaScript and Python (12).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-ergonomics):

- 0 to 25, context cost. For MCP, the number and size of the tool definitions (25 for ten or fewer compact tools, 15 for 11 to 30, 5 for more than 30, plus up to 10 back for toolsets, dynamic loading or read-only subsets). For APIs, whether responses can be sized (field selection, limits, summaries).
- 20, pagination, filtering and output-size controls.
- 20, actionable, documented error responses, codes and messages an agent can recover from.
- 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations.
- 15, sensible defaults, few required parameters, and official SDKs in at least two languages.

Models are read for tool use, structured output, prompt caching, context length, batch and SDKs. Frameworks for how much code and how many defaults a tool-calling agent with MCP needs.

## 4. Schema & documentation, 73 out of 100, up to 4.4 more on the total

Why it scored 73: The Management API has an OpenAPI 3.1 schema in beta, but the Authentication API, where the token exchange happens, has none (10). llms.txt for the agent docs at auth0.com/ai/docs/llms.txt with 67 links (10). The agent docs explain when to use Token Vault, CIBA or FGA and what each is for (16). The token exchange reference lists required and optional parameters with their fixed URN values, and marks the scope subset as Early Access (12). One success example and the 401 and 403 cases on the exchange reference, which is thin for an error contract (10). Dated public changelog and semver SDKs (15).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-schema):

APIs and MCP servers.

- 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool).
- 10, llms.txt or Markdown docs served for agents.
- 0 to 20, descriptions that say what a tool is for, when to use it and when not to, read from the tool definitions in the source or the API reference.
- 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs.
- 0 to 15, examples and documented error responses.
- 15, versioning and a public changelog.

Models are read from the API reference, the OpenAPI file, llms.txt, the structured-output and tool-use docs and the model cards. Frameworks from docs a model can follow, typed interfaces, examples and the API reference.

## 5. Maintenance & community, 74 out of 100, up to 2.3 more on the total

Why it scored 74: Latest dated changelog entry 18 September 2026, with Custom Token Exchange GA on 28 August (30). Well over three dated changelog entries in the last 90 days (20). Closed platform with a public changelog and a community forum, while auth0-ai-js has a bug report from April 2025 with no visible fix (10). @auth0/ai 6.0.2 shipped on 22 April 2026 and the Python auth0-ai 1.0.2 on 20 January 2026, so the agent SDKs are five to eight months old (8). An open request from 13 September 2026 asks to move LangChain from 0.3 to 1.0 (6).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-maintenance):

- 0 to 30, time since the last release, or the last published model or API change for a closed service. 30 within 30 days, 20 within 90, 10 within 180, 0 older.
- 20, at least three releases or dated changelog entries in the last 90 days.
- 0 to 25, responsiveness. Issues and pull requests answered on GitHub (the open issues and how recent the replies are). For closed services, a public changelog and a support or community channel that answers, 0 to 15.
- 15, presence in the official MCP registry under a verified namespace (MCP servers), or current official SDKs (APIs and models).
- 10, package health, current dependencies and CI.

Models are read for deprecation notice periods and model churn rather than release counts.

## 6. Security & auth, 88 out of 100, up to 2.1 more on the total

Why it scored 88: OAuth 2.0 and OIDC with RFC 8693 token exchange, DPoP binding and scoped provider tokens, less a little because the refresh-token exchange needs refresh token rotation turned off (28). CIBA with RAR asks the user to approve the exact action on a second device, a scope subset can be requested and FGA filters what a RAG agent can read (20). Token Vault returns tokens, not untrusted content (10). Tenant logs stream to Datadog, Splunk, EventBridge and others, but retention is 1 day on Free and 5 on Essentials (10). Valid security.txt, Bugcrowd programmes for Okta and Auth0, and SDK advisories published on GitHub (20).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-security):

- 0 to 30, the credential model. 30 for OAuth 2.1 with scopes, or scoped and revocable keys with rotation. 20 for plain revocable API keys. 10 for one all-powerful key. 10 off when a secret can travel in a URL query string as a documented option.
- 0 to 20, read-only or least-privilege modes, and confirmation or approval for destructive actions.
- 0 to 15, prompt-injection posture where the tool returns untrusted content (documented mitigations or guidance). A tool that returns no untrusted content gets 10.
- 0 to 15, audit logs or per-call visibility for the operator.
- 0 to 20, a security programme. security.txt or a disclosure policy, a bug bounty, SOC 2 or ISO 27001, advisories handled in public.

Models are read for retention, whether API data trains models (and whether that's off by default), zero-retention options and certifications. Frameworks for telemetry defaults, approval hooks, guardrails and sandboxing.

## 7. Transparency & trust, 85 out of 100, up to 1.3 more on the total

Made of editorial 70, provenance 100.

Why it scored 85: Closed platform under Okta's terms with Apache-2.0 SDKs (15). Okta publishes a privacy policy, a DPA and a subprocessor list, and Auth0 states log retention per plan, but we couldn't read the subprocessor page on 2026-10-01 to check it against Auth0's hosting regions (20). A deprecations page lists each change with announcement and end-of-life dates six to seven months apart (20). Okta publishes a subprocessor list, but we couldn't open it ourselves and didn't check Auth0's hosting regions this run (15).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-transparency):

- 0 to 30, source availability and licence clarity. 30 for open source under an OSI licence, 15 for closed with clear terms, 0 for unclear terms.
- 0 to 30, data handling and retention statements that agree with each other (privacy policy, DPA, retention periods, subprocessors).
- 0 to 20, a deprecation policy or notices with dates.
- 0 to 20, telemetry disclosed with an opt-out (local software), or subprocessors and data locations disclosed (hosted).

The other half of Transparency and trust is the provenance score, computed from checked facts (below). The category score is the mean of the two.

## What we couldn't check

What we couldn't read counted as absent. Publishing it on a page a plain HTTP fetch can read (not only in a browser) lets the next check count it.

- Auth0's own incident history for July to September 2026 didn't render, so the reliability record rests on third-party trackers we haven't confirmed.
- Whether the 50 per cent agent add-on is sold self-serve, since the public pricing page names AI identity add-ons only under Enterprise.
- We couldn't open Okta's subprocessor page on 2026-10-01.

## Weaknesses

- Only works when Auth0 is the identity provider for your users
- Two Token Vault connections on Free and three on Essentials and Professional without the add-on
- Log retention of 1 day on Free and 5 days on Essentials is short for an audit trail
- No OpenAPI schema for the Authentication API that the exchange uses
- Agent SDKs last released in April 2026 (JavaScript) and January 2026 (Python)

## What costs an agent a turn today

The notes we give agents before they call it. Each one is a workaround an agent shouldn't need.

- Turn off refresh token rotation on the application before using the refresh token exchange
- Treat a 401 from the exchange as a missing connected account and send the user through the Connected Accounts flow
- Pass login_hint when a user has linked two accounts from the same provider
- Use CIBA for purchases or deletes and wait for the approval instead of asking in chat
- Read X-RateLimit-Reset on a 429 and back off until then

## What the review panel asked for

- Fewer dashboard-only steps
- exchange with rotation on
- longer log retention

## When it's done

Send what changed and where it's published as a dispute (https://www.anchorterminal.com/builders/#disputes, or `POST https://www.anchorterminal.com/api/v1/contact` with `"kind": "dispute"`). Disputes are answered in public, and the listing is checked again by the same checklist. Paying for an audit or a listing claim changes nothing here.