Bria API

# Fix list: Bria API

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/bria, the October 2026 research run, assessed 8 October 2026. Grade B, 63.7 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 Bria API: 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. Security & auth, 33 out of 100, up to 11.7 more on the total

Why it scored 33: Several named, revocable organisation keys with no scopes, plus OAuth 2.1 with PKCE on the hosted MCP server, where the scopes are identity only (20). Less 10 because the streaming background removal WebSocket takes `api_token` as a URL query parameter (10 net). No read-only or restricted key and no spend cap found in the reviewed documentation (0). Returns generated images and a structured prompt built from the caller's own input (10). Each response has a `request_id`. Audit logs are listed for Enterprise only (5). The site states ISO 27001 certification and SOC 2 attestation with no report, auditor or trust centre linked. No `security.txt`, disclosure policy or bug bounty found (8). The no-training warranty is counted under Transparency.

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.

## 2. Payments & pricing, 40 out of 100, up to 7.5 more on the total

Why it scored 40: No x402, MPP or L402 (0). Per-image, per-call and per-second prices published without a login (20). 100 free calls with no card (20). A person has to sign up in the console to get a key. The MCP OAuth flow has dynamic client registration but still needs a Bria account sign-in (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.

## 3. Transparency & trust, 49 out of 100, up to 4.5 more on the total

Made of editorial 42, provenance 56.

Why it scored 49: Closed hosted API under the Online AI General Terms and Conditions of March 2025, a PDF linked from the website terms. SDKs and the MCP server are MIT. Model weights are source-available for non-commercial use (15). The terms warrant no training on customer content without written consent and deletion within 30 days of termination. The docs give 3 days for image results and say inputs are not kept beyond processing. The privacy policy has no date and no retention periods, and does not name Google, whose Gemini 2.5 Flash handles prompts on the standard generate endpoint (20). No deprecation policy. One endpoint is marked deprecated without a date (3). The privacy policy lists recipient categories only, with vendor names on request by email. Data locations not stated (4).

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.

Provenance checks not met in full (half of this category, computed from checked facts):

- Domain age: bria.ai, registered 2020-04-03 (6 years) (11 of 15)
- Endpoint on the vendor's domain: engine.prod.bria-api.com is not on bria.ai (0 of 15)
- Terms of service: published, but our reader couldn't read it (7 of 10)
- Privacy policy: read, states 6 of the 8 things a reader expects (8.5 of 10)
- Changelog: not found (0 of 10)
- security.txt: not found (0 of 10)

## 4. Agent ergonomics, 76 out of 100, up to 3.9 more on the total

Why it scored 76: Results come back as URLs, with a `seed` and the full structured prompt. The MCP server has 18 tools and returns a JPEG preview of up to 0.7 MB inline (18). Resolution, aspect ratio, output type and a sync or async choice, with nothing to paginate (15). One error shape with `request_id`, a status table and messages the MCP server passes through as `Bria API error <status code>: <reason>` (18). No idempotency key. Webhooks carry `Bria-Webhook-Id` for deduplication. The MCP tools carry annotations, but all 18 are marked read-only and idempotent although each call is billed (10). Only a prompt is required, and there are official Python and TypeScript SDKs (15).

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.

## 5. Reliability, 83 out of 100, up to 3.4 more on the total

Why it scored 83: Read with the hosted lines. Atlassian Statuspage at status.bria.ai with API components and history (20). One incident between 10 July and 8 October 2026, an API outage for image editing, increase resolution and image generation on 22 September, labelled major and resolved in 18 minutes with no cause given. Under an hour, so minor by our line (20). Limits published per plan, 10, 60 and 1,000 requests a minute per endpoint (15). The docs tell callers to back off with jitter and honour `Retry-After` on 429, and the SDKs retry status polls only so a job is not started twice. No idempotency key (13). The site states a 99.9 per cent uptime SLA for managed cloud, but no SLA document was found and the online terms supply the products as is, so half credit (5). The REST API is generally available (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.

## 6. Maintenance & community, 69 out of 100, up to 2.7 more on the total

Why it scored 69: `bria-mcp` 2.2.1 on 4 October 2026, four days before the check (30). For models we look at churn and notice. FIBO, FIBO Lite and the v1 product endpoints are all still served and one v1 endpoint is marked deprecated with its v2 replacement named, but no notice policy or dated notice was found (10). No API changelog. Support is by email and a Discord community, not tested, and the GitHub API refused us so SDK issues were not read (6). Current official SDKs, `bria-client` 0.3.0 on PyPI (24 June 2026) and `@bria-ai/client` 0.2.0 on npm (28 September 2026). The MCP server is not in the official MCP registry (15). The SDK repository has CI, release automation and a lock file. The SDKs are 0.x and `bria-mcp` is classified beta (8).

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.

## 7. Schema & documentation, 84 out of 100, up to 2.6 more on the total

Why it scored 84: A downloadable OpenAPI 3.0 file for each API, such as `/_bundle/image-generation.yaml` (25). `llms.txt`, a Markdown twin of every page and a docs MCP server at docs.bria.ai/mcp (10). Endpoint pages state the purpose and the valid input combinations, and the MCP tool descriptions say when to use `edit_image` in place of `text_to_image` (17). Enums for resolution, aspect ratio and output type, but `structured_prompt` is a JSON document passed as a string (12). Request and response examples, an error table and typed error bodies per endpoint, with `error.code` shown only as a placeholder (13). Paths are versioned v1 and v2, but the OpenAPI files have an empty version and no API changelog was found (7).

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.

## 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.

- The lead was right on the interface (REST API, Python and TypeScript SDKs, official MCP server). The TypeScript SDK was first published on 28 September 2026.
- unchecked: GitHub stars and open issues for Bria-AI/bria-client. The GitHub API answered with a rate limit, so `githubStars` is empty and responsiveness is scored without the issue tracker.
- unchecked: which terms the console presents at sign-up. The Online AI General Terms and Conditions are linked from the website terms for demo tools and define APIs as Online Services, so we treat them as governing.
- unchecked: the hosted MCP server's live tool list and annotations. The count of 18 and the annotations come from the `bria-mcp` 2.2.1 source, and the docs say the hosted server exposes the same capabilities.
- The pricing page lists video background removal at $0.0225 a second and bria.ai/llms.txt at $0.01. We used the pricing page.
- No SLA document was found behind the 99.9 per cent uptime statement.
- No effective date on the privacy policy, and no API launch date found, so `firstReleased` is empty.
- Whether failed or moderated requests are billed is not stated in the reviewed pages.

## Weaknesses

- API keys are organisation-wide with no scopes, and no spend cap was found in the reviewed documentation
- No API changelog or deprecation policy was found, and the OpenAPI files carry an empty version
- The streaming background removal WebSocket takes `api_token` as a URL query parameter
- No `security.txt`, disclosure policy or bug bounty was found. ISO 27001 and SOC 2 are stated on the site with no report or trust centre linked
- All 18 tools in `bria-mcp` are marked read-only and idempotent although each call is billed

## 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.

- Send the key in the `api_token` header, not as `Authorization: Bearer`. Only `/v2/image/verify` and `/v2/video/verify` work without it
- Expect `202` with `request_id` and `status_url` from v2 endpoints. Poll `GET /v2/status/{request_id}` every 1 to 2 seconds, or pass `sync: true`
- Status polls count towards the per-endpoint limit (10 a minute on Free, 60 on Pay-as-you-go), so prefer `webhook_url` for batches
- Download results promptly. Image URLs expire after 3 days by default, and legacy `/v1/product/*` results after 1 hour
- Do not resend a failed `POST` blindly. There is no idempotency key, and the SDKs retry status polls only

## 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.