Built for Windsurf Cascade
Sprint Tickets,
In Cascade Mode.
Windsurf's Cascade agent makes multi-file edits look easy. Wire SprintFlint in and Cascade pulls the next ticket, refactors across files, and posts the result.
5-min setup
REST API + MCP server
No card to start
Why Windsurf + SprintFlint
Built around how your AI actually works.
Other trackers
- Cascade can't see your sprint without browser scraping
- No agent-friendly ticket API
- Multi-file edits stall when context is missing
- Per-seat pricing punishes the agent seat
SprintFlint
- Public MCP server at mcp.sprintflint.com; native protocol
- Bearer-token auth, one header, no OAuth dance
- REST endpoints designed for agent-first workflows
- £5/user/mo, free tier covers your first 300 tickets
Step 1: Token
Token in your env.
export SPRINTFLINT_TOKEN=sf_pk_your_token_here
# Smoke test from terminal
curl -H "Authorization: Bearer $SPRINTFLINT_TOKEN" \
https://sprintflint.com/api/v1/dashboard
Step 2: .windsurfrules
Add the Windsurf rule.
Drop a rule file at the project root. Cascade now reads SprintFlint context whenever it kicks off multi-file work.
# .windsurfrules
When starting work on a SprintFlint ticket, GET
https://sprintflint.com/api/v1/issues/$TICKET first to load
full description and acceptance criteria.
On finish, POST a comment to /api/v1/comments summarising
the changes and tests added.
Step 3: Run
Cascade ships the ticket.
"Cascade, take the next SprintFlint ticket." Cascade fetches it, plans the multi-file edit, ships, and comments.
- Agent reads sprint backlog
- Auto-updates status when work begins
- Posts progress comments as it goes
- Pairs with Autoplay for full one-click PRs
Other AI editors we support
Same setup, different tools.
Cascade Was Built for Big Edits.
Pair it with a sprint board that keeps up.
300 tickets free. No card. Live in 5 minutes.