Built for terminal-first agents

Sprint Tickets,
Claude-Native.

Pair SprintFlint with Claude Code and your agent reads the sprint, picks the next ticket, ships the PR, and comments back. All from your terminal.

Agent skill ready
Public /llms.txt
Pairs with Autoplay

Why Claude Code + SprintFlint

Designed for agents that read code and ship.

Other tools

  • API requires OAuth dances and SDK installs
  • No public agent-readable spec
  • UIs assume a human is clicking
  • Per-seat creep — you pay for the agent's seat too

SprintFlint

  • Bearer-token auth, single header, zero SDK
  • /llms.txt + REST so Claude discovers everything
  • Endpoints shaped for agent loops (list → act → comment)
  • £5/user/mo, agents share your seat — no double-billing
Step 1 — Token

One header, full access.

# export the token once
export SPRINTFLINT_TOKEN=sf_pk_your_token_here

# smoke test from your terminal
curl -H "Authorization: Bearer $SPRINTFLINT_TOKEN" \
  https://sprintflint.com/api/v1/dashboard
Step 2 — Drop in the skill

Teach Claude Code the workflow.

Save the SprintFlint agent skill so Claude Code knows when and how to read tickets, branch, ship, and comment.

# ~/.claude/skills/sprintflint/SKILL.md
---
name: sprintflint
description: Read SprintFlint sprint context and post progress.
  Use when the user asks about tickets, velocity, or sprints.
---

When the user asks about their sprint:
1. GET /api/v1/dashboard — fetch active sprint + assigned tickets
2. Pick highest-priority undone ticket
3. Implement, commit, push
4. POST /api/v1/comments with PR link + summary
Step 3 — Run

Hand off the sprint.

"Claude, ship the next SprintFlint ticket."
Reads the backlog. Picks one. Branches. Codes. Tests. Comments. Done.

  • Agent reads assigned tickets
  • Branches and PRs from terminal
  • Posts back to the right ticket
  • Pairs with SprintFlint Autoplay for fully autonomous loops

Your Terminal Already Ships Code.

Now your sprint board can keep up.
300 tickets free. No credit card. Set up before your next standup.