Guide · Agent skill
How to use the Sprint Status Snapshot agent skill
Updated · written with AI from the project's own documentation and checked against our catalogue data
The sprint-status skill is a small instruction package for Claude that gives a fast, read-only snapshot of how your sprint is going: time left, which stories are complete, in progress, blocked, or not started, and at most one recommendation. This guide explains what an agent skill is, how to install the sprint-status ZIP, how to ask for a snapshot in plain language, and what the skill needs and does on your machine.
Quick start
- 1Download the ZIP. Download the sprint-status ZIP from this page. It contains SKILL.md.
- 2Choose an install location. Use ~/.claude/skills/ for personal use or a project's .claude/skills/ folder for one project.
- 3Unzip the skill. Unzip the package so that SKILL.md sits inside a sprint-status folder in the skills directory.
- 4Check your project layout. Confirm the project has production/sprints/ files and story files, and optionally production/sprint-status.yaml.
- 5Ask a sprint question. Type a request such as 'How is the sprint going?' or give a sprint number.
- 6Read the snapshot. Review time remaining, story statuses, any STALE flags and the single recommendation.
What the sprint-status skill does
When you ask how the sprint is going, the agent reads your sprint plan and story files, works out how much time is left in the sprint, and checks which stories are complete, in progress, blocked, or not started. The result is a short summary of about 30 lines.
The skill is strictly read-only. It does not edit files or propose changes, and it makes at most one concrete recommendation. It is a situational check, not a sprint review or planning session.
- Reads the sprint plan and story files in production/sprints and the story files it references.
- Calculates days elapsed, days remaining and percentage of time used.
- Reports status for each story and flags stories with no recent progress as stale.
- If production/sprint-status.yaml exists, uses it as the authoritative status source.
What is an agent skill?
An agent skill is a folder containing a SKILL.md file. The file tells Claude when to use the skill, which tools it may use, and what steps to follow. Claude loads the skill when your request matches its description, so you can ask in ordinary language instead of memorising a command.
The sprint-status skill is built for a game-development workflow that uses a project folder layout with production/sprints and story files. It is not useful in a project without that layout.
Install the sprint-status skill
Download the ZIP for the sprint-status skill from this page. The package contains SKILL.md. Unzip it into one of two places depending on how you want to use it.
For personal use across projects, unzip into ~/.claude/skills/. For a single project, unzip into that project's .claude/skills/ folder. You can also upload the ZIP in the Claude apps under Settings → Capabilities → Skills.
- Personal install: ~/.claude/skills/sprint-status/SKILL.md
- Project install: .claude/skills/sprint-status/SKILL.md inside your project
- Claude apps: Settings → Capabilities → Skills, then upload the ZIP
Ask for a sprint snapshot with natural requests
The skill is triggered by questions about sprint progress. The skill's own example is 'How is the sprint going?' You can also pass an optional sprint number as an argument, and leave it blank to use the current sprint.
If you give a sprint number, the skill looks for a matching file such as sprint-003.md, sprint-03.md or sprint-3.md in production/sprints/ and reports which file it found. Without a number, it uses the most recently modified file in that folder.
- How is the sprint going?
- Give me a sprint snapshot.
- What is left in the current sprint and is anything blocked?
- Sprint status for sprint 3.
Read the snapshot
The output covers the sprint goal and dates, time consumed, story status and, where relevant, an Attention Needed section. Stories flagged stale have had no progress in more than 4 days, based on a Last Updated style date in the story file. Stories with no date are reported as never stamped rather than stale.
If any in-progress story is stale, the burndown verdict is raised to at least At Risk, even when completion is within the normal range.
- Complete and in-progress counts, plus blocked and not-started stories.
- STALE flags for in-progress stories with no update in more than 4 days.
- One recommendation at most.
What the skill needs and its safety notes
The skill needs a Bash shell for its helper scripts, yaml-helper.sh and story-status.sh, and the Claude Code runtime with the allowed tools Read, Glob, Grep and Bash. Your project should contain production/sprints/ files, story files, and optionally production/sprint-status.yaml.
The skill runs scripts. It does not use the network. Its tool permissions are limited to reading files, searching, and the specific Bash script calls listed in the skill. Review the SKILL.md file before installing if you want to confirm exactly what runs.
- Runs scripts: yes (yaml-helper.sh and story-status.sh)
- Uses network: no
- Read-only: it does not edit files
Tips for better results
Keep sprint files and story files in the expected layout, and keep story status fields current. The skill reads the status you have written, so stale or missing status produces a less useful snapshot.
For planning or a full review, use the commands the skill itself points to rather than this snapshot.
- Add a Last Updated line to story files so staleness can be checked.
- Create production/sprint-status.yaml when you can, since it is the authoritative source.
- Use /sprint-plan and /milestone-review for planning and full reviews.
What people use it for
- A solo developer checking which story to start next from a one-page build order.
- A lead running a quick situational check before deciding whether to use /sprint-plan or /milestone-review.
- A game team using the Claude Code Game Studios structure who wants a fast progress check at the start of a day.
- Spotting in-progress stories that have not been updated for more than 4 days.
Frequently asked questions
How do I install the sprint-status skill?
Download the ZIP from this page and unzip it into ~/.claude/skills/ for personal use, or into a project's .claude/skills/ folder. You can also upload the ZIP in the Claude apps under Settings → Capabilities → Skills.
Is the sprint-status skill free and what is its licence?
The source project is licensed under MIT. The source does not mention pricing, so check the directory listing for any cost details.
What do I need to run sprint-status?
You need a Bash shell for its helper scripts, the Claude Code runtime with Read, Glob, Grep and Bash tools, and a project containing production/sprints/ files and story files. production/sprint-status.yaml is optional.
Does sprint-status change my files?
No. The skill is strictly read-only. It does not edit files or propose changes, and it makes at most one recommendation.
Why does it say my project has no sprint files?
The skill expects a production/sprints/ folder with sprint files. Without that layout it reports that files are missing. At the minimal workflow setting, it reads story files instead and reports build order progress.
How is sprint-status different from sprint planning tools?
It is only a snapshot. For full sprint reviews or planning, the skill points you to /sprint-plan and /milestone-review.
New agent skills, repos and AI tools — every week
The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.
Our verdict
Use sprint-status if you work in the Claude Code Game Studios structure, or in a project with production/sprints and story files, and want a fast, read-only progress check. Pick something else if you need full sprint reviews or planning, or if your project does not follow that folder layout.
Download the skill