Get one useful thing working before you configure everything.
Start with one real product question. Specky helps you turn it into something you can review.
Try saying
“Here is what we are hearing from customers. What is the clearest problem to solve first?”
Bring one thing
A product URL, note, or customer quote.
Ask clearly
Say what you are trying to decide.
Review together
Check the evidence before you act.
Specky API reference
What this is for
The Specky API lets your code and your agents work with one Specky workspace: search customer evidence, read product context, capture signals, and draft documents, decisions, and tickets for a person to review.
Get a key
Create a workspace API key in Settings → Developer. Send it on every request:
curl https://www.specky.space/api/v1/search?q=onboarding -H "Authorization: Bearer sk_live_your_key"
A key belongs to one workspace, and it can only reach that workspace. Reading works on any workspace. Creating work and AI operations need a paid plan or an active trial, and trials need a card up front. See pricing.
Full specification
The OpenAPI 3.1 specification lists every endpoint, parameter, and response field. Point your code generator or agent at it.
Common endpoints
| Job | Request |
|---|---|
| Search evidence and documents | GET /api/v1/search?q=… |
| Read confirmed product context | GET /api/v1/product-context |
| Get a brief before coding | GET /api/v1/coding-brief?q=… |
| Save customer feedback | POST /api/v1/capture |
| Add signals in bulk | POST /api/v1/signals |
| Draft a document | POST /api/v1/documents |
| Draft tickets | POST /api/v1/tickets |
| See what awaits review | GET /api/v1/inbox |
Successful tool calls return ok: true, a readable summary, and the structured fields listed in the specification.
Errors
Every 4xx and 5xx response is JSON with two fields you can rely on:
{ "error": "Invalid or revoked token.", "code": "unauthorized" }
Use code in your program and show error to people. Common codes are invalid_request, unauthorized, payment_required, forbidden, not_found, conflict, rate_limited, and internal_error.
Rate limits
You can make 120 requests per minute per key, or per IP address without a key. AI operations also count against your plan's limits. Every response tells you where you stand:
| Header | Meaning |
|---|---|
RateLimit-Limit | Requests allowed per window |
RateLimit-Remaining | Requests left in this window |
RateLimit-Reset | Seconds until the window resets |
Retry-After | Seconds to wait, sent with a 429 |
The same values are available as X-RateLimit-Limit, X-RateLimit-Remaining, and X-RateLimit-Reset, where the reset is a Unix timestamp.
Other ways to connect
- Specky MCP server for Claude, Cursor, and other MCP clients.
- Specky CLI for terminals and CI jobs.
- Versioning and deprecation policy for how the API changes.
Ready to apply this?
Start with your own product and keep the first read grounded. You'll create an account before the guided result so it can be saved to your workspace.