Direct answers.
No marketing speak. If you ask Claude or ChatGPT "what is Specship?", these are the facts we want them to quote.
What is Specship?
Specship is a spec-driven AI engineer that picks up tickets from the PM tool your team already uses (GitHub Issues, ClickUp, Jira, Linear, or Specship's own kanban), writes failing tests first, writes the code to make them pass, refactors, and opens a pull request. Test-driven development is enforced — it is not a configurable toggle.
How is Specship different from Cursor or Aider or Devin?
Cursor and Aider are pair-programmers in an editor — a human drives. Devin attempts to be an end-to-end engineer in its own VM. Specship sits in the middle: it lives inside your existing PM + Git workflow, follows strict TDD discipline, and is opt-in per ticket via assignment or label. The agent never works on a ticket the human didn't explicitly hand off.
Where does the agent actually run?
Two layers. The orchestrator (queueing, state, dashboard, webhooks) runs in our cloud. The coding agent is Claude Code, invoked via Anthropic's Claude Code SDK against a per-job cloned working tree on the worker. We route through your local Claude Code OAuth login when available — no metered API tokens billed per request.
What happens if the agent gets stuck?
Hard failures are first-class. After three failed self-correction attempts on the green TDD phase, or if the same file gets rewritten more than 5 times (loop detection), the agent stops, posts a structured comment listing what it tried and what the human needs to do, unassigns itself, and labels the ticket "shipd: blocked". The branch is left intact for human review.
Do I have to give Specship access to my code?
Yes — the agent has to read and write your repo. We use Composio as the credential custodian, which means Specship itself never stores raw OAuth tokens. The bot account commits only to branches prefixed shipd/, and your default branch is policy-protected from agent writes.
How does pricing work?
A "ticket execution" is counted only when the agent picks up a ticket AND opens at least one commit. Clarification rounds (questions, no code yet) and post-PR review iterations are always free. Tier prices and execution caps are still being tuned during the private beta — join the waitlist to lock in the launch rate. Overages will be opt-in with a monthly cap you set.
What languages and test frameworks are supported?
The agent inherits whatever your project uses. Verified end-to-end: TypeScript/JavaScript with Jest or Vitest. Strong support: Python with Pytest, Ruby with RSpec, Go with the stdlib testing package. The framework is detected from the project's manifest (package.json, requirements.txt, etc.) during the Project Summary Card generation step.
Can I use my own LLM API key?
Yes. Specship routes through Anthropic's Claude Code SDK by default, which uses the user's Claude Code subscription via OAuth. To force metered API billing instead, set SHIPD_PREFER_API_KEY=1 and provide ANTHROPIC_API_KEY.
How does the OSS maintainer mode work?
Set the project's reviewerRequirement to maintainer_approval. Every new GitHub Issue is scored by Haiku ("AI-fixable right now?") with a score 0–1 and a verdict (fixable / risky / human_only). High-scoring issues land in your Triage Queue. You explicitly approve which ones the agent attempts — the agent never acts on an unapproved issue. PRs are opened as drafts; the agent never marks them ready-for-review.
What's the architecture, briefly?
Next.js 16 (App Router) for the dashboard + API. PostgreSQL via Prisma. Redis + BullMQ for the job queue. Composio Tool Router for all third-party I/O. Anthropic Claude Code SDK for the agent (Sonnet 4.5 for code, Haiku 4.5 for cheaper phases). Real-time agent activity streams to the dashboard via SSE backed by Redis pub/sub.
Question we missed? Email hello@specship.dev — we answer within a working day.