Every API here is available over the APIs.io API and to AI agents over MCP.
MCP server
One button, every client — Claude, Cursor, VS Code and the rest.
https://apis.io/mcp
Tools for apis
7 MCP tools reach this
find_apisBrowse and filter every API in the catalog.
get_api_artifactsOne API's artifacts, grouped by type.
get_openapiThe primary OpenAPI for this API.
find_similar_apisAPIs that look like this one.
apis_io_searchSTART HERE — APIs, providers and tags for one query, each with its total.
resolveTurn a domain, URL or GitHub org into the provider it belongs to.
find_cohortsEvery scored population of providers in the catalog.
All 92 tools →
Call it yourself
curl for this page
This API
curl "https://apis.io/api/v1/apis/buildpulse-io:buildpulse-io-submissions-api"
All apis
curl "https://apis.io/api/v1/apis?limit=25"
Calls without X-API-Key need no key. Calls with it need a paid plan — get a key.
Get an API key
Free tier, no form to fill in. Signing in shares your email address with us — we
store it to create your key and to recognise you if you sign in with another
provider. See our Privacy Policy and
Terms.
A second provider on the same verified email joins the account you already have.
openapi: 3.2.0
info:
title: BuildPulse Platform Submissions API
version: 1.0.0
summary: Public REST API for BuildPulse CI analytics.
termsOfService: https://buildpulse.io/terms
description: 'The BuildPulse Platform API exposes read-only test analytics for any
repository onboarded to BuildPulse: paginated test submissions,
disruptive/flaky test inventory, per-test disruption history, and SVG
badges for flakiness and coverage. This API is the drop-in replacement for the legacy Rails API at
`https://buildpulse.io/api`. Endpoint paths, query parameters, and
response JSON shapes are intentionally identical so existing
integrations migrate by changing only the base URL. ## Base URLs
- Production: `https://platform.buildpulse.io`
- Development: `https://platform.dev.buildpulse.io`
## Authentication
All endpoints except `/health` require an API token in the
`Authorization` header. Both `token ` and `Bearer `
prefixes are accepted. Two token shapes are accepted:
- **New** (`bp_`) — minted by the BuildPulse web app under
Organization Settings → API Tokens. SHA256-hashed (lowercase hex)
and stored in `apiTokens` as
`{ type: "api_token", tokenPrefix, tokenHash, expiresAt }`. Plan-gated: the organization needs an active plan that includes
API access (Team, Growth, custom, or a paid legacy Marketplace
plan; the 14-day trial counts). The same gate applies to the
hosted MCP''s `mcpSessions`. - **Legacy** (``) — Rails-era format, stored as
`{ hashedToken: base64(SHA256(token)) }`. Gated against
`marketplacePurchases.unitCount > 0` for parity with the legacy
API. Production currently holds zero of these; the path is
retained for any future migrated tokens. The plaintext token is shown only once at creation time. Example values (do not use):
```
Authorization: Bearer bp_14f56e0c2a8b4d1f9e6a3c7b8d2e5f1a0b9c4d7e6f3a2b1c0d9e8f7a6b5c4d3e
Authorization: token a1b2c3d4e5f6a7b8c9d0e1f2a3b4c5d6e7f8a9b0
```
## Response shapes
JSON timestamps are ISO 8601 with second precision in UTC
(e.g. `2026-05-17T12:34:56Z`). List endpoints always return an array
for the data field, never `null`. Empty results return `[]`. ## Errors
| Status | Body |
|--------|------|
| 401 | empty |
| 404 | `{"message":"Not Found"}` |
| 400 | `{"errors":{"after":[""]}}` |
| 500 | empty |
The empty-body 401/500 contract matches the legacy Rails behavior
(`render body: nil`) and is preserved deliberately. ## Pagination
Two cursor styles are in use, both base64-encoded so consumers should
treat them as opaque:
- **Keyset (submissions only)** — cursor is `base64("id:")`. Stable under inserts. - **Offset (tests, flaky tests)** — cursor is `base64("offset:")`. The next-page cursor is returned as `metadata.after`. When there is
no next page, `metadata.after` is `null`. ## Reporting window
Time-bounded endpoints use a rolling window ending now. Which window
depends on the endpoint:
| Endpoint | Window |
|---|---|
| `GET /api/v1/flaky/tests?quarantine=false` | Your plan''s reporting period (default 14 days) |
| `GET /api/tests/{id}/results` | Your plan''s reporting period (default 14 days) |
| `GET /api/v1/flaky/tests` (quarantined view, the default) | Not windowed — every quarantined test is returned |
| `GET /api/repos/{owner}/{name}/tests` | Not windowed — every test with a disruptiveness ratio above 0 |
| `GET /api/repos/{owner}/{name}/submissions` | Fixed 14 days |
| `GET /api/v1/flaky/badges` | Fixed 14 days |
The plan''s reporting period is resolved per organization from its
active plan, so it follows plan changes automatically; there is no
request parameter to override it. Organizations without a resolvable
plan get the 14-day default. Don''t hardcode a window client-side —
treat the returned rows as the source of truth. ## For AI agents (MCP)
The BuildPulse MCP server exposes this API as nine read-only tools
for Claude, ChatGPT, Cursor, VS Code, and any Model Context
Protocol client. Hosted: `https://mcp.buildpulse.io/mcp`
(Streamable HTTP, OAuth sign-in with your BuildPulse account). Stdio: `npx -y @buildpulse/mcp` with `BUILDPULSE_TOKEN` set to an
API token. Setup and tool list:
https://platform.buildpulse.io/docs/mcp.'
license:
name: Proprietary
url: https://buildpulse.io/terms
contact:
name: BuildPulse Support
email: support@buildpulse.io
url: https://buildpulse.io
servers:
- url: https://platform.buildpulse.io
description: Production
- url: https://platform.dev.buildpulse.io
description: Development
security:
- TokenAuth: []
- BearerAuth: []
tags:
- name: Submissions
description: Test result sets (a.k.a. submissions) — one per CI run.
paths:
/api/repos/{owner}/{name}/submissions:
get:
operationId: listSubmissions
tags:
- Submissions
summary: List recent test submissions for a repository
description: 'Returns the last 14 days of test result sets ("submissions") for
the named repository, newest first. Pagination is keyset-based on
the document ID — `metadata.after` is an opaque cursor that
should be passed back as the `after` query parameter to fetch the
next page.
Repository lookup is case-insensitive; `OWNER/REPO` matches
`owner/repo`.'
parameters:
- $ref: '#/components/parameters/Owner'
- $ref: '#/components/parameters/Name'
- $ref: '#/components/parameters/Limit10'
- $ref: '#/components/parameters/AfterCursor'
responses:
'200':
description: Paginated submissions.
content:
application/json:
schema:
$ref: '#/components/schemas/SubmissionsResponse'
example:
count: 2
submissions:
- key: 1234/5678/test-results.xml.gz
build_url: https://github.com/acme/widgets/actions/runs/1234
commit_oid: a1b2c3d4e5f6a7b8c9d0e1f2a3b4c5d6e7f8a9b0
recorded_at: '2026-05-17T12:34:56Z'
status: processed
test_result_count: 412
- key: 1234/5679/test-results.xml.gz
build_url: https://github.com/acme/widgets/actions/runs/1235
commit_oid: b2c3d4e5f6a7b8c9d0e1f2a3b4c5d6e7f8a9b0a1
recorded_at: '2026-05-17T11:00:00Z'
status: processed
test_result_count: 412
metadata:
after: aWQ6NjVkNGYxYThiM2MyZDFlMGY5YThiN2M2
limit: 10
'400':
$ref: '#/components/responses/BadCursor'
'401':
$ref: '#/components/responses/Unauthorized'
'402':
$ref: '#/components/responses/PlanLimitExceeded'
'404':
$ref: '#/components/responses/NotFound'
x-codeSamples:
- lang: shell
label: curl
source: "curl -H \"Authorization: token $BUILDPULSE_TOKEN\" \\\n \"https://platform.buildpulse.io/api/repos/acme/widgets/submissions?limit=5\"\n"
- lang: javascript
label: Node
source: "const res = await fetch(\n \"https://platform.buildpulse.io/api/repos/acme/widgets/submissions?limit=5\",\n { headers: { Authorization: `token ${process.env.BUILDPULSE_TOKEN}` } }\n);\nconst { submissions, metadata } = await res.json();\n"
- lang: python
label: Python
source: "import os, requests\nr = requests.get(\n \"https://platform.buildpulse.io/api/repos/acme/widgets/submissions\",\n headers={\"Authorization\": f\"token {os.environ['BUILDPULSE_TOKEN']}\"},\n params={\"limit\": 5},\n)\nr.raise_for_status()\ndata = r.json()\n"
/api/repos/{owner}/{name}/submissions/{submission_id}/tests:
get:
operationId: listSubmissionTests
tags:
- Submissions
summary: List per-test results for a single submission
description: 'Returns the individual test results recorded against one
submission (one CI run). Use `status=failed` to filter to just
the failures/errors — useful when triaging a red test run.
The submission identifier is the 24-char hex `id` returned by
`GET /api/repos/{owner}/{name}/submissions`. Each row carries
`conclusion` (`passing`/`failure`/`error`/`skipped`),
`duration_us` (per-attempt microseconds), `message` (failure
message, may be null), and the parent test case''s name, suite,
file, and classname.
Repository lookup is case-insensitive; `OWNER/REPO` matches
`owner/repo`. The submission must belong to a repository the
caller''s token can access — otherwise a 404 is returned (no
information is leaked about cross-tenant submissions).'
parameters:
- $ref: '#/components/parameters/Owner'
- $ref: '#/components/parameters/Name'
- name: submission_id
in: path
required: true
schema:
type: string
description: Submission identifier (24-char hex ObjectID).
- name: status
in: query
required: false
schema:
type: string
enum:
- all
- failed
- passed
- failure
- error
- skipped
description: 'Optional conclusion filter. `failed` matches both
`failure` and `error` (the typical "red test run" set).
Omit or pass `all` to return every result.
'
- $ref: '#/components/parameters/Limit25'
- $ref: '#/components/parameters/AfterCursor'
responses:
'200':
description: Paginated test results for the submission.
content:
application/json:
schema:
$ref: '#/components/schemas/SubmissionTestsResponse'
example:
count: 3
tests:
- id: 65d4f1a8b3c2d1e0f9a8b7c6
test_case_id: 65d4f1a8b3c2d1e0f9a8b7c0
name: logs in with valid credentials
suite: Authentication
class: AuthSpec
file: spec/auth_spec.rb
conclusion: failure
duration_us: 1245000
message: expected 200 but got 500
ran_at: '2026-05-27T18:42:11Z'
run_count: 3
metadata:
after: null
limit: 25
'400':
$ref: '#/components/responses/BadCursor'
'401':
$ref: '#/components/responses/Unauthorized'
'402':
$ref: '#/components/responses/PlanLimitExceeded'
'404':
$ref: '#/components/responses/NotFound'
x-codeSamples:
- lang: shell
label: curl
source: "curl -H \"Authorization: token $BUILDPULSE_TOKEN\" \\\n \"https://platform.buildpulse.io/api/repos/acme/widgets/submissions/65d4f1a8b3c2d1e0f9a8b7c6/tests?status=failed\"\n"
components:
schemas:
NotFoundError:
type: object
required:
- message
properties:
message:
type: string
enum:
- Not Found
Submission:
type: object
required:
- id
- key
- build_url
- commit_oid
- recorded_at
- status
- test_result_count
properties:
id:
type: string
description: 'Submission identifier (24-char hex ObjectID). Pass to
`/api/repos/{owner}/{name}/submissions/{submission_id}/tests`
to fetch the per-test results for this submission.
'
example: 65d4f1a8b3c2d1e0f9a8b7c6
key:
type: string
description: 'Legacy S3 object key for the raw test result archive. Stable
and unique per submission in the legacy stack; empty string
on the new pipeline. Prefer `id` for cross-endpoint joins.
'
example: 1234/5678/test-results.xml.gz
build_url:
type: string
format: uri
description: CI run URL that produced this submission.
commit_oid:
type: string
description: Git commit SHA-1 (40 hex chars).
recorded_at:
type: string
format: date-time
description: ISO 8601 timestamp (UTC, second precision).
status:
type: string
description: 'Processing status. Currently `processed`, `processing`, or
other internal states.
'
example: processed
test_result_count:
type: integer
description: 'Number of individual test results in this submission. Counts every
recorded attempt — retries and skipped tests included — so it can
exceed the number of distinct tests in the run.
'
SubmissionTest:
type: object
required:
- id
- test_case_id
- name
- suite
- class
- file
- conclusion
- duration_us
- message
- body
- ran_at
- run_count
properties:
id:
type: string
description: Test result identifier (24-char hex ObjectID).
test_case_id:
type: string
description: 'Parent test case identifier (24-char hex ObjectID). Two
results with the same `test_case_id` are different runs of
the same logical test (e.g. retries within one submission).
'
name:
type: string
suite:
type: string
class:
type: string
file:
type: string
conclusion:
type: string
enum:
- passing
- failure
- error
- skipped
- unknown
description: 'Outcome of this attempt. `failed` in the `status` filter
matches both `failure` and `error`.
'
duration_us:
type: integer
format: int64
description: Per-attempt duration in microseconds.
message:
type:
- string
- 'null'
description: 'One-line failure or error summary as recorded by the runner;
null for passing / skipped.
'
body:
type:
- string
- 'null'
description: 'Full failure detail (assertion + stack trace); null for
passing / skipped. Truncated at 2000 bytes on the wire
(UTF-8 safe, never split mid-codepoint); the authoritative
copy lives in the underlying testResults row.
'
ran_at:
type: string
format: date-time
run_count:
type: integer
description: 'Retry attempt number recorded by the test runner. `1` for
the first attempt, `2` for the first retry, etc.
'
PlanLimitError:
type: object
required:
- code
- message
- upgradeUrl
properties:
code:
type: string
enum:
- plan_limit_exceeded
description: Stable machine-readable code; consumers key off this, not the message.
message:
type: string
description: Human-readable explanation.
upgradeUrl:
type: string
format: uri
description: Deep link to billing/upgrade so access can be restored.
SubmissionTestsResponse:
type: object
required:
- count
- tests
- metadata
properties:
count:
type: integer
format: int64
description: Total matching test results (not just this page).
tests:
type: array
items:
$ref: '#/components/schemas/SubmissionTest'
metadata:
$ref: '#/components/schemas/Metadata'
CursorError:
type: object
required:
- errors
properties:
errors:
type: object
required:
- after
properties:
after:
type: array
items:
type: string
SubmissionsResponse:
type: object
required:
- count
- submissions
- metadata
properties:
count:
type: integer
format: int64
description: Total submissions matching the filter (not just this page).
submissions:
type: array
items:
$ref: '#/components/schemas/Submission'
metadata:
$ref: '#/components/schemas/Metadata'
Metadata:
type: object
required:
- after
- limit
properties:
after:
type:
- string
- 'null'
description: Opaque cursor for the next page; `null` if there is no next page.
limit:
type: integer
parameters:
Owner:
in: path
name: owner
required: true
schema:
type: string
description: Repository owner (case-insensitive).
example: acme
Limit25:
in: query
name: limit
schema:
type: integer
minimum: 1
maximum: 100
default: 25
description: Page size. Clamped to `[1, 100]`.
Name:
in: path
name: name
required: true
schema:
type: string
description: Repository name (case-insensitive).
example: widgets
Limit10:
in: query
name: limit
schema:
type: integer
minimum: 1
maximum: 100
default: 10
description: Page size. Clamped to `[1, 100]`.
AfterCursor:
in: query
name: after
schema:
type: string
description: 'Opaque pagination cursor from a previous response''s
`metadata.after`. Pass back verbatim.
'
responses:
BadCursor:
description: Cursor failed to decode.
content:
application/json:
schema:
$ref: '#/components/schemas/CursorError'
example:
errors:
after:
- invalid cursor format
PlanLimitExceeded:
description: 'The organization is over its plan''s monthly test-result limit and
data-read access is restricted (staged-overage stage `soft` or
`hard`). At `soft` the data is locked but safe — ingestion keeps
recording. At `hard` ingestion is also paused (raw retained ~30
days) and an upgrade resumes processing. Read `GET /api/v1/usage`
for the current stage and usage, and visit `upgradeUrl` to restore
access. Only returned for organizations positively identified as
over the enforced limit while enforcement is enabled.
'
content:
application/json:
schema:
$ref: '#/components/schemas/PlanLimitError'
example:
code: plan_limit_exceeded
message: Your organization has exceeded its plan's monthly test-result limit. Your data is safe but locked — upgrade your plan to restore access.
upgradeUrl: https://buildpulse.io/settings/billing
Unauthorized:
description: 'Missing, malformed, or invalid token; or — for legacy 40-hex
tokens only — the organization has no active plan
(`marketplacePurchases.unitCount > 0`). The new `bp_<64-hex>`
path has no plan gate, so a 401 there means the token shape
was wrong, the hash didn''t match, or the token has expired.
Response body is empty (matches legacy Rails).
'
NotFound:
description: Resource not found.
content:
application/json:
schema:
$ref: '#/components/schemas/NotFoundError'
example:
message: Not Found
securitySchemes:
TokenAuth:
type: apiKey
in: header
name: Authorization
description: "Send `Authorization: token <token>`. Two token shapes are\naccepted:\n\n- **New** (`bp_<64-hex>`) — minted by the refactored web-client\n today (`web-client/src/app/_services/apiTokens.ts`). Stored as\n `{ type: \"api_token\", tokenPrefix, tokenHash (hex SHA256),\n expiresAt }`. No active-plan gate; issuance is the\n entitlement check.\n- **Legacy** (`<40-hex>`) — Rails-era format, stored as\n `{ hashedToken: base64(SHA256(token)) }`. Plan-gated against\n `marketplacePurchases.unitCount > 0`. Retained for any future\n migrated tokens.\n\nThe legacy API only accepted `token`, so existing integrations\nneed no changes.\n"
BearerAuth:
type: http
scheme: bearer
bearerFormat: bp_<64-hex> or <40-hex>
description: 'Send `Authorization: Bearer <token>`. Functionally identical to
`TokenAuth`; both header forms are accepted. Same two token
shapes apply — see `TokenAuth` for the format and plan-gate
details.
'
externalDocs:
description: BuildPulse docs, MCP server and guides
url: https://platform.buildpulse.io/docs/mcp