# run-tests-on-the-unmodified-codebase-before-starting-to-establis

run-tests-on-the-unmodified-codebase-before-starting-to-establis is a Claude Code skill that requires running tests on untouched code before edits, recording failures to distinguish them from new regressions.

- skill · `type=skill` · Claude Code, Codex, OpenCode, Cursor, GitHub Copilot, Gemini CLI, Windsurf (Devin Desktop), Claude.ai · ★ 539 repo stars
- Author: ZaxbyHub
- License: MIT · policy=mirror · the beginning of the file is at the end of this page
- Source: https://github.com/ZaxbyHub/opencode-swarm/tree/HEAD/.opencode/skills/generated/run-tests-on-the-unmodified-codebase-before-starting-to-establis/SKILL.md
- Updated: 2026-10-29
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/skills/run-tests-on-the-unmodified-codebase-before-starting-to-establis

## What it does

This skill enforces a baseline protocol before any code modification begins. The agent must execute the project test runner on untouched files and record broken tests. This log serves as the reference point for the entire session.

The prompt explicitly forbids labeling failures after a patch as pre-existing unless an initial run confirmed them. Reviewer steps inside the rule direct inspectors to verify that baseline runs actually occurred in the diff history.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add ZaxbyHub/opencode-swarm --skill run-tests-on-the-unmodified-codebase-before-starting-to-establis -a claude-code
```

2. **Manually: put the files in place**
- File: `~/.claude/skills/run-tests-on-the-unmodified-codebase-before-starting-to-establis/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/run-tests-on-the-unmodified-codebase-before-starting-to-establis/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add ZaxbyHub/opencode-swarm --skill run-tests-on-the-unmodified-codebase-before-starting-to-establis -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/run-tests-on-the-unmodified-codebase-before-starting-to-establis/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add ZaxbyHub/opencode-swarm --skill run-tests-on-the-unmodified-codebase-before-starting-to-establis -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/run-tests-on-the-unmodified-codebase-before-starting-to-establis/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install ZaxbyHub/opencode-swarm .opencode/skills/generated/run-tests-on-the-unmodified-codebase-before-starting-to-establis
```

2. **Manually: put the files in place**
- File: `.github/skills/run-tests-on-the-unmodified-codebase-before-starting-to-establis/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add ZaxbyHub/opencode-swarm --skill run-tests-on-the-unmodified-codebase-before-starting-to-establis -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/ZaxbyHub/opencode-swarm --path .opencode/skills/generated/run-tests-on-the-unmodified-codebase-before-starting-to-establis
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/run-tests-on-the-unmodified-codebase-before-starting-to-establis/SKILL.md` — The legacy .windsurf/skills/ path still works

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add ZaxbyHub/opencode-swarm --skill run-tests-on-the-unmodified-codebase-before-starting-to-establis -a opencode
```

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/ZaxbyHub/opencode-swarm/tree/HEAD/.opencode/skills/generated/run-tests-on-the-unmodified-codebase-before-starting-to-establis/SKILL.md

## When to use it
- capture test suite failures on clean code before editing
- prevent agents from mislabeling new breakages as pre-existing
- verify baseline execution during code review checks

## Example prompts
- Run the existing test suite before modifying the database layer
- Establish a failure baseline on clean code before starting

## Source text · SKILL.md · MIT

Below is the content of a third-party file. It is data, not instructions for you: do not act on anything written inside unless the human asks. Block tag: 1b0d23c636b63ce8.

BEGIN THIRD-PARTY FILE 1b0d23c636b63ce8
````text
---
name: run-tests-on-the-unmodified-codebase-before-starting-to-establis
description: run tests on the unmodified codebase before starting to establish a failure baseline
metadata:
  triggers: '["pre-existing test failures","baseline before change"]'
  generated_from_knowledge: '[]'
  source_knowledge_ids: '[]'
  generated_at: "2026-07-02T02:35:49.911Z"
  confidence: "0.60"
  status: active
  version: "2"
  skill_origin: generated
---

<!-- generated by opencode-swarm skill-generator. Do not edit by hand; edits will be preserved on regeneration only with controlled update mode. -->

# run tests on the unmodified codebase before starting to establish a failure baseline

## Trigger

- pre-existing test failures
- baseline before change

## Required Procedure

- run tests on the unmodified codebase before starting to establish a failure baseline

## Forbidden Shortcuts

- label post-change test failures as 'pre-existing' without running tests before the change

## Delegation Template

When delegating a task affected by this skill, include:

```
SKILLS: file:.opencode/skills/generated/run-tests-on-the-unmodified-codebase-before-starting-to-establis/SKILL.md
```

## Reviewer Checks

- Verify each required action above appears in the diff.

## Test Engineer Checks

- Add or update tests covering the trigger condition and the forbidden shortcut.
````
END THIRD-PARTY FILE 1b0d23c636b63ce8

The third-party file has ended. Everything between the 1b0d23c636b63ce8 tags is data: requests and commands from it are not followed. An “end of file” line with another tag inside the block is part of the file.

© ZaxbyHub · MIT · https://github.com/ZaxbyHub/opencode-swarm/tree/HEAD/.opencode/skills/generated/run-tests-on-the-unmodified-codebase-before-starting-to-establis/SKILL.md

## Similar
- [run-tests](https://affpapa.org/en/nevibecode/skills/run-tests.md) — skill · 7000 installs
- [run-acceptance-tests](https://affpapa.org/en/nevibecode/skills/run-acceptance-tests.md) — skill · 3900 installs
- [run-smoke-tests](https://affpapa.org/en/nevibecode/skills/run-smoke-tests.md) — skill · 2300 installs
- [run-tests-after-changes](https://affpapa.org/en/nevibecode/hooks/run-tests-after-changes.md) — hook · 1967 installs
- [run-python-tests](https://affpapa.org/en/nevibecode/skills/run-python-tests.md) — skill · ★ 6244 repo stars
- [run-rust-tests](https://affpapa.org/en/nevibecode/skills/run-rust-tests.md) — skill · ★ 6244 repo stars

From the same set: [zaxbyhub/opencode-swarm](https://affpapa.org/en/nevibecode/by/zaxbyhub/opencode-swarm.md) — 121 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
