agentleFS
Sign inStart free
ArangoGutierrezArangoGutierrez/claude-toolkit/.cursor/rules/tdd.mdc

claude-toolkit / rules

Cursor rule4 starsChanged 3 months agoMITView on GitHub

TDD protocol for implementing features and fixing bugs — RED/GREEN/REFACTOR phase signals, checkpoint gating, context verification

We split this tdd.mdc into the rules it gives an agent

7 rules. About 0 are common, 6 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Before any REFACTOR touching > 3 files or > 50 LOC:

    line 21
  2. Distinctive:

    Ask human to review diff and commit

    line 23
  3. Distinctive:

    Wait for confirmation before proceeding

    line 24
  4. Distinctive:

    If any missing → ask before writing code.

    line 32
  5. Distinctive:

    Run language-appropriate scans before claiming implementation is complete:

    line 35
  6. Distinctive:

    Use the language-specific rule (go.mdc, etc.) for exact scan commands

    line 37
  7. Very specific:

    Always run: trivy fs . for filesystem/dependency vulnerability scan

    line 38

≈ “Common” is an estimate: at least 1 in 5 of the 6,350 Cursor rule files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
tdd.mdc
---
description: TDD protocol for implementing features and fixing bugs — RED/GREEN/REFACTOR phase signals, checkpoint gating, context verification
alwaysApply: false
---
# TDD Protocol

## Phase Signals (prefix every implementation response)
- `[PLAN]` — designing/specifying before code
- `[RED]` — writing or presenting a failing test
- `[GREEN]` — writing minimum code to pass the test
- `[REFACTOR]` — cleaning up after green, no behavior change
- `[CHECKPOINT]` — requesting human commit before large refactor

## Hard Rules
- RED before GREEN: never write implementation without a failing test first
- One phase per turn: never mix test changes and implementation in the same edit
- Fitness function: tests are contracts — never modify tests to make implementation pass
- Batch size: each PR ≤ 1 concern; break down first if > 1

## Checkpoint Protocol
Before any REFACTOR touching > 3 files or > 50 LOC:
1. Signal `[CHECKPOINT]` and summarize current GREEN state
2. Ask human to review diff and commit
3. Wait for confirmation before proceeding
4. Begin REFACTOR as a new atomic change after checkpoint

## Context Verification
At implementation start, verify you have:
- API docs for external services being integrated
- Team conventions for the language/framework in use
- Existing test patterns in the repo
If any missing → ask before writing code. Never guess at APIs or conventions.

## Security Scans (verify phase)
Run language-appropriate scans before claiming implementation is complete:
- Detect project language from files present (go.mod → Go, package.json → Node, etc.)
- Use the language-specific rule (go.mdc, etc.) for exact scan commands
- Always run: `trivy fs .` for filesystem/dependency vulnerability scan

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free