# AI use cases for Finn Loop (finna/Finn-loop)

Repo: [finna/Finn-loop](https://github.com/finna/Finn-loop) | Use cases: 4 | Page: https://dedupedai.com/use-cases/finna-finn-loop/

Finn-loop is a set of three Claude Code skills that use Linear and GitHub to run a small, human-gated loop from spec to build to review, with a human making the merge.

This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL.

- Turn a rough idea into an agent-ready Linear issue: https://dedupedai.com/use-cases/finna-finn-loop/spec-idea-into-agent-ready-issue/
- Have Claude build queued issues into pull requests: https://dedupedai.com/use-cases/finna-finn-loop/build-agent-ready-issues-into-prs/
- Get a verdict on every open PR before merging: https://dedupedai.com/use-cases/finna-finn-loop/review-open-prs-before-merging/
- Install the three Finn Loop skills into a GitHub repo: https://dedupedai.com/use-cases/finna-finn-loop/install-finn-loop-skills-in-repo/

## How to turn a rough idea into an agent-ready Linear issue with Finn Loop

Repo: [finna/Finn-loop](https://github.com/finna/Finn-loop) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README

Interview with /finn-spec until the behavior is clear, then get a Linear issue with acceptance criteria and non-goals.

**Who it is for:** A developer or founder with a feature idea for a GitHub repo who wants it written up as a build-ready contract before any code is written.

**The problem:** Vague feature requests produce confident but wrong pull requests, and writing precise acceptance criteria takes time.

**The outcome:** A Linear issue with observable acceptance criteria (AC-N) and binding non-goals (NG-N), labeled agent-ready only after the human approves it.

### Steps

1. Open Claude Code inside the target repo and run `/finn-spec`.
2. Answer the interview questions until the behavior is unambiguous; the skill researches the repo first.
3. Read the filed Linear issue and check each AC-N and NG-N against what you want.
4. If you approve the exact contract, apply the agent-ready label in Linear yourself.

### Prompt to paste into your agent

```text
In the GitHub repo at <repo path>, run the /finn-spec skill for this idea: <idea text>. Expected output: one Linear issue with AC-N acceptance criteria and NG-N non-goals. Verify that every acceptance criterion describes an observable outcome before reporting the issue link.
```

### Caveat

The README says spec quality is the bottleneck, so the interview is only as good as the answers you give it.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/finna-finn-loop/spec-idea-into-agent-ready-issue/

## How to have Claude build queued issues into pull requests with Finn Loop

Repo: [finna/Finn-loop](https://github.com/finna/Finn-loop) | Difficulty: Intermediate | Time: 45 minutes | Basis: From the README

Run a builder loop that claims the next safe agent-ready issue, implements only its contract, and opens a PR.

**Who it is for:** A developer with a Linear team that has agent-ready issues waiting and wants an agent to work through them one at a time.

**The problem:** Small, well-specified tasks sit in a backlog because they take attention that is hard to spare during the day.

**The outcome:** One pull request per issue, each tied to its Linear contract, ready for review.

### Steps

1. Confirm the Linear issue is labeled agent-ready and not blocked, and is unassigned.
2. Start the builder in Claude Code with `/loop /finn-build` and keep the session open.
3. Watch the first few passes and your usage before leaving it running.
4. Check each opened PR against its Linear issue before anything else happens.

### Prompt to paste into your agent

```text
In the GitHub repo at <repo path>, run /loop /finn-build against the Linear team <team key>. Expected output: a PR linked to the claimed issue. Verify that the PR changes only what that issue's acceptance criteria require and respects its non-goals.
```

### Caveat

Run only one builder loop per Linear team, because two simultaneous sessions authenticated as the same person cannot reliably lock each other.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/finna-finn-loop/build-agent-ready-issues-into-prs/

## How to get a verdict on every open PR before merging with Finn Loop

Repo: [finna/Finn-loop](https://github.com/finna/Finn-loop) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README

Run a reviewer loop that checks open PRs against their Linear issue and required checks, then posts a verdict.

**Who it is for:** A developer who has agent-built PRs waiting and needs a consistent first-pass review before deciding whether to merge.

**The problem:** Reading every agent PR in full against its issue and CI status takes time, and it is easy to miss a missing acceptance criterion.

**The outcome:** Each open PR gets a three-group verdict. Only PRs marked loop-approved, conflict-free, and green on all required checks are candidates for the human merge.

### Steps

1. Start a second Claude Code session in the repo and run `/loop /finn-review`.
2. Read the verdict each PR receives, including any loop-changes-requested or needs-human-review label.
3. Merge only PRs that are loop-approved, conflict-free, and green on all required checks.
4. For any needs-human-review PR, read and resolve the escalation reason before merging.

### Prompt to paste into your agent

```text
In the GitHub repo at <repo path>, run /loop /finn-review on the open pull requests. Expected output: a three-group verdict posted on each PR with its Finn-loop label. Verify that each verdict names the linked Linear issue and the required checks it was measured against.
```

### Caveat

loop-approved is evidence for the human merge decision, not permission to merge, and fully automated verdicts need at least one required GitHub status check.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/finna-finn-loop/review-open-prs-before-merging/

## How to install the three Finn Loop skills into a GitHub repo with Finn Loop

Repo: [finna/Finn-loop](https://github.com/finna/Finn-loop) | Difficulty: Intermediate | Time: 60 minutes | Basis: From the README

Copy the finn-spec, finn-build, and finn-review skills into a repo and check that the prerequisites are working.

**Who it is for:** A developer setting up Finn Loop for the first time in a GitHub repo with a Linear team who wants the skills available in Claude Code.

**The problem:** The skills need the right Claude Code version, a working gh login, Linear labels, and a team key before they can run.

**The outcome:** The three skills appear in /skills, and a smoke test confirms Linear and GitHub reads succeed.

### Steps

1. Check the Claude Code version with `claude --version` and confirm it is 2.1.71 or newer.
2. Check GitHub access with `gh auth status` and `gh repo view`.
3. Paste the install prompt from the README into Claude Code inside the repo, and give it your Linear team key when asked.
4. Run `/reload-skills` (or restart Claude Code), then run `/skills` and confirm finn-spec, finn-build, and finn-review are listed.

### Prompt to paste into your agent

```text
Inside the GitHub repo at <repo path>, follow the Finn-loop install steps from the README of finn/Finn-loop using the Linear team key <team key>. Expected output: the three SKILL.md files under .claude/skills and a smoke-test report. Verify that /skills lists finn-spec, finn-build, and finn-review.
```

### Caveat

The install prompt tells the agent to detect the repo's real default branch and confirm the authenticated account can push, so verify those results before trusting the setup.

Risk, credentials: The GitHub CLI must be authenticated with write access to the target repo, so the account used can push code changes there.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/finna-finn-loop/install-finn-loop-skills-in-repo/