Agent skill

Iterative Development

by alinaqi in alinaqi/maggy

TDD iteration loops using Claude Code Stop hooks - runs tests after each response, feeds failures back automatically

MITAuto-check passedTesting & QA

Install Iterative Development

skills CLI
$ npx skills add alinaqi/maggy --skill iterative-development -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install alinaqi/maggy iterative-development --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/alinaqi/maggy.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/iterative-development .claude/skills/iterative-development && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
iterative-development
GitHub stars
707
Token cost
~2.2k tokens
SKILL.md length
334 words
Files
1
Skills in repo
71
Repo updated
First seen
Licence
MIT

At a glance

TDD iteration loops using Claude Code Stop hooks - runs tests after each response, feeds failures back automatically

  • Works in 3 steps: Remove or rename the Stop hook in… → Or set MAX_ITERATIONS=1 in the script → Or delete scripts/tdd-loop-check.sh
  • Tasks that involve Test-driven development
  • SKILL.md covers How It Actually Works, Setup: Stop Hook Configuration, Additional Hooks for Quality… and Core Philosophy, plus 4 more sections
  • Calls npm, npx and pytest

What it does

Iterative Development is an agent skill from alinaqi/maggy. TDD iteration loops using Claude Code Stop hooks - runs tests after each response, feeds failures back automatically

Its SKILL.md is about 2.2k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Testing & QA, covering Test-driven development. The repository describes itself as: What started as an opinionated Claude Code setup kit is now an autonomous AI engineering command center. The licence is MIT.

When your agent uses it

  • Tasks that involve Test-driven development

Example prompts

  • “/iterative-development”

Requirements

  • Python 3
  • Node.js

Workflow steps

3 steps, taken from the first numbered list in SKILL.md.

  1. Remove or rename the Stop hook in .claude/settings.json
  2. Or set MAX_ITERATIONS=1 in the script
  3. Or delete scripts/tdd-loop-check.sh

What it can do on your machine

Read from SKILL.md and the folder at commit 72a456e. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • npm
    • npx
    • pytest
    • ruff
    • mypy

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use npm and npx, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Iterative Development loads about 2.2k tokens when it runs. Until then it costs about 35 tokens; SKILL.md has 334 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~35
When it runs · the whole SKILL.md, loaded when a task matches
~2.2k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from alinaqi/maggy at commit 72a456e, republished under its MIT licence (© alinaqi). 334 words, ~2,201 tokens.

Download SKILL.mdSave it as .claude/skills/iterative-development/SKILL.md (or your agent's skills folder).
name
iterative-development
description
TDD iteration loops using Claude Code Stop hooks - runs tests after each response, feeds failures back automatically
when-to-use
When setting up or configuring TDD loops via Stop hooks
user-invocable
false
effort
medium

Iterative Development Skill (Stop Hook TDD Loops)

Concept: Claude Code's Stop hook fires right before Claude finishes a response. Exit code 2 feeds stderr back to the model and continues the conversation. This creates a real TDD loop without any plugins.


How It Actually Works

Claude Code has a Stop hook that runs when Claude is about to conclude its response. If the hook script exits with code 2, its stderr is shown to the model and the conversation continues automatically.

┌─────────────────────────────────────────────────────────────┐
│  1. User asks Claude to implement a feature                 │
├─────────────────────────────────────────────────────────────┤
│  2. Claude writes tests + implementation                    │
├─────────────────────────────────────────────────────────────┤
│  3. Claude finishes its response                            │
├─────────────────────────────────────────────────────────────┤
│  4. Stop hook runs: executes tests, lint, typecheck         │
├─────────────────────────────────────────────────────────────┤
│  5a. All pass (exit 0) → Claude stops, work is done         │
│  5b. Failures (exit 2) → stderr fed back to Claude          │
├─────────────────────────────────────────────────────────────┤
│  6. Claude sees failures, fixes code, response ends         │
├─────────────────────────────────────────────────────────────┤
│  7. Stop hook runs again → repeat until green or max tries  │
└─────────────────────────────────────────────────────────────┘

Key insight: No fake plugins, no /ralph-loop command. The hook is real Claude Code infrastructure that runs automatically.


Setup: Stop Hook Configuration

Add this to your project's .claude/settings.json:

json
{
  "hooks": {
    "Stop": [
      {
        "hooks": [
          {
            "type": "command",
            "command": "scripts/tdd-loop-check.sh",
            "timeout": 60,
            "statusMessage": "Running tests..."
          }
        ]
      }
    ]
  }
}
The TDD Loop Check Script

Create scripts/tdd-loop-check.sh in your project:

bash
#!/bin/bash
# TDD Loop Check - runs after each Claude response
# Exit 0 = all good, Claude stops
# Exit 2 = failures, stderr fed back to Claude to fix

MAX_ITERATIONS=25
ITERATION_FILE=".claude/.tdd-iteration-count"

# Track iteration count
if [ -f "$ITERATION_FILE" ]; then
    count=$(cat "$ITERATION_FILE")
    count=$((count + 1))
else
    count=1
fi
echo "$count" > "$ITERATION_FILE"

# Safety: stop after max iterations
if [ "$count" -ge "$MAX_ITERATIONS" ]; then
    rm -f "$ITERATION_FILE"
    echo "Max iterations ($MAX_ITERATIONS) reached. Stopping loop." >&2
    exit 0
fi

# Skip if no test files exist yet
if ! find . -name "*.test.*" -o -name "*.spec.*" -o -name "test_*" 2>/dev/null | grep -q .; then
    rm -f "$ITERATION_FILE"
    exit 0
fi

# Run tests
TEST_OUTPUT=$(npm test 2>&1) || {
    echo "ITERATION $count/$MAX_ITERATIONS - Tests failing:" >&2
    echo "$TEST_OUTPUT" | tail -30 >&2
    echo "" >&2
    echo "Fix the failing tests and try again." >&2
    exit 2
}

# Run lint (if configured)
if [ -f "package.json" ] && grep -q '"lint"' package.json; then
    LINT_OUTPUT=$(npm run lint 2>&1) || {
        echo "ITERATION $count/$MAX_ITERATIONS - Lint errors:" >&2
        echo "$LINT_OUTPUT" | tail -20 >&2
        echo "" >&2
        echo "Fix lint errors and try again." >&2
        exit 2
    }
fi

# Run typecheck (if configured)
if [ -f "tsconfig.json" ]; then
    TYPE_OUTPUT=$(npx tsc --noEmit 2>&1) || {
        echo "ITERATION $count/$MAX_ITERATIONS - Type errors:" >&2
        echo "$TYPE_OUTPUT" | tail -20 >&2
        echo "" >&2
        echo "Fix type errors and try again." >&2
        exit 2
    }
fi

# All green - reset counter and let Claude stop
rm -f "$ITERATION_FILE"
exit 0
Python Variant
bash
#!/bin/bash
# Python TDD Loop Check

MAX_ITERATIONS=25
ITERATION_FILE=".claude/.tdd-iteration-count"

if [ -f "$ITERATION_FILE" ]; then
    count=$(cat "$ITERATION_FILE")
    count=$((count + 1))
else
    count=1
fi
echo "$count" > "$ITERATION_FILE"

if [ "$count" -ge "$MAX_ITERATIONS" ]; then
    rm -f "$ITERATION_FILE"
    echo "Max iterations ($MAX_ITERATIONS) reached." >&2
    exit 0
fi

if ! find . -name "test_*" -o -name "*_test.py" 2>/dev/null | grep -q .; then
    rm -f "$ITERATION_FILE"
    exit 0
fi

TEST_OUTPUT=$(pytest -v 2>&1) || {
    echo "ITERATION $count/$MAX_ITERATIONS - Tests failing:" >&2
    echo "$TEST_OUTPUT" | tail -30 >&2
    exit 2
}

if command -v ruff &>/dev/null; then
    LINT_OUTPUT=$(ruff check . 2>&1) || {
        echo "ITERATION $count/$MAX_ITERATIONS - Lint errors:" >&2
        echo "$LINT_OUTPUT" | tail -20 >&2
        exit 2
    }
fi

if command -v mypy &>/dev/null; then
    TYPE_OUTPUT=$(mypy . 2>&1) || {
        echo "ITERATION $count/$MAX_ITERATIONS - Type errors:" >&2
        echo "$TYPE_OUTPUT" | tail -20 >&2
        exit 2
    }
fi

rm -f "$ITERATION_FILE"
exit 0

Additional Hooks for Quality Enforcement

PreToolUse Hook: Lint Before File Writes

Runs a linter before any Write/Edit lands:

json
{
  "hooks": {
    "PreToolUse": [
      {
        "matcher": "Write|Edit",
        "hooks": [
          {
            "type": "command",
            "command": "scripts/pre-write-lint.sh",
            "timeout": 10,
            "statusMessage": "Checking code quality..."
          }
        ]
      }
    ]
  }
}
SessionStart Hook: Auto-Inject Context

Runs at session start to inject project info:

json
{
  "hooks": {
    "SessionStart": [
      {
        "hooks": [
          {
            "type": "command",
            "command": "echo 'TDD loop active. Tests run automatically after each response. Fix failures to continue.'",
            "statusMessage": "Loading project context..."
          }
        ]
      }
    ]
  }
}

Core Philosophy

┌─────────────────────────────────────────────────────────────┐
│  ITERATION > PERFECTION                                     │
│  ─────────────────────────────────────────────────────────  │
│  Don't aim for perfect on first try.                        │
│  Let the loop refine the work. Each iteration builds on     │
│  previous attempts visible in files and git history.        │
├─────────────────────────────────────────────────────────────┤
│  FAILURES ARE DATA                                          │
│  ─────────────────────────────────────────────────────────  │
│  Failed tests, lint errors, type mismatches are signals.    │
│  The Stop hook feeds them directly to Claude as context.    │
├─────────────────────────────────────────────────────────────┤
│  CLEAR COMPLETION CRITERIA                                  │
│  ─────────────────────────────────────────────────────────  │
│  The hook defines "done": tests pass, lint clean, types ok. │
│  No ambiguity about when to stop.                           │
└─────────────────────────────────────────────────────────────┘

Error Classification

Not all failures should loop. The hook script should distinguish:

TypeExamplesAction
Code ErrorLogic bug, wrong assertion, type mismatchExit 2 → loop continues
Access ErrorMissing API key, DB connection refusedExit 0 → stop, report to user
Environment ErrorMissing package, wrong runtime versionExit 0 → stop, report to user

The sample scripts above handle this — they only exit 2 for test/lint/type failures, not for environment issues.


When to Use TDD Loops

Good For
Use CaseWhy
Feature developmentTests provide clear pass/fail signal
Bug fixesWrite failing test, fix, loop until green
RefactoringExisting tests catch regressions
API developmentEach endpoint independently testable
Not Good For
Use CaseWhy
UI/UX workRequires human judgment
One-shot operationsNo iteration needed
Unclear requirementsNo clear "done" criteria
Subjective designNo objective success metric

Disabling the Loop

To temporarily disable the TDD loop for a session:

  1. Remove or rename the Stop hook in .claude/settings.json
  2. Or set MAX_ITERATIONS=1 in the script
  3. Or delete scripts/tdd-loop-check.sh

The hook only fires if the script exists and is configured.


Gitignore Additions

gitignore
# TDD loop state
.claude/.tdd-iteration-count

© alinaqi, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in skills/iterative-development of alinaqi/maggy.

Open the folder on GitHubat commit 72a456e

Compare with similar skills

Iterative Development next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Iterative Development compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Iterative Development this skillalinaqi/maggy707—~2.2kAutomated safety check: PassMIT
TDDpietheinstrengholt/rssmonster56430 repos~906Automated safety check: PassMIT
TDD WorkflowhellangleZ/burn-in-cceverywhere-ralph11211 repos~2.4kAutomated safety check: PassNone
TDDsanity-io/sanity6.4k20 repos~1kAutomated safety check: PassMIT
Test Driven Developmentfarm-fe/farm5.6k51 repos~2.5kAutomated safety check: PassMIT
Tapd Story PipelineTencentBlueKing/bk-bcs840—~2.6kAutomated safety check: PassCustom licence

Similar skills

  • TDD

    pietheinstrengholt/rssmonster

    Test-driven development. An agent skill from pietheinstrengholt/rssmonster.

    564 GitHub starsUsed in 30 repos~906 tokens
    Testing & QAAuto-check passed
  • TDD Workflow

    hellangleZ/burn-in-cceverywhere-ralph

    A skill your agent uses when writing new features, fixing bugs, or refactoring code.

    112 GitHub starsUsed in 11 repos~2.4k tokens
    Testing & QAAuto-check passed
  • TDD

    sanity-io/sanity

    Official

    Test-driven development with red-green-refactor loop. An agent skill from sanity-io/sanity.

    6.4k GitHub starsUsed in 20 repos~1k tokens
    Testing & QAAuto-check passed
  • A skill your agent uses when implementing any feature or bugfix, before writing implementation code

    5.6k GitHub starsUsed in 51 repos~2.5k tokens
    Testing & QAAuto-check passed
  • Tapd Story Pipeline

    TencentBlueKing/bk-bcs

    单需求实现流水线——把一个 TAPD 需求从零推进到代码提交。自动串联技术澄清、 开发计划、任务拆分、TDD 实现、架构/安全校验、代码提交六个阶段。

    840 GitHub stars~2.6k tokensUpdated today
    Testing & QAAuto-check passed
  • Absolute Init

    maddhruv/absolute

    One-time setup for absolute: interview how you want it to behave (output style, autonomy, TDD strictness, spec dir, families) + detect the stack once, then write .absolute.config.json (project…

    218 GitHub starsUsed in 1 repo~3k tokens
    Testing & QAAuto-check passed

More from alinaqi/maggy

All 71 skills in this repo
  • AI Models

    alinaqi/maggy

    Latest AI models reference - Claude, OpenAI, Gemini, Eleven Labs, Replicate

    707 GitHub starsUsed in 1 repo~4.1k tokens
    Auto-check passed
  • Azure Cosmosdb

    alinaqi/maggy

    Azure Cosmos DB partition keys, consistency levels, change feed, SDK patterns

    707 GitHub starsUsed in 1 repo~4.5k tokens
    Auto-check passed
  • LLM Patterns

    alinaqi/maggy

    AI-first application patterns, LLM testing, prompt management

    707 GitHub starsUsed in 1 repo~2.1k tokens
    Auto-check passed
  • Woocommerce

    alinaqi/maggy

    WooCommerce REST API - products, orders, customers, webhooks

    707 GitHub starsUsed in 1 repo~4.4k tokens
    Auto-check: notes
  • Aeo Optimization

    alinaqi/maggy

    AI Engine Optimization - semantic triples, page templates, content clusters for AI citations

    707 GitHub stars~3.7k tokensUpdated 14 days ago
    Auto-check passed
  • Agent Teams

    alinaqi/maggy

    Claude Code Agent Teams - default team-based development with strict TDD pipeline enforcement

    707 GitHub stars~5k tokensUpdated 14 days ago
    Auto-check: notes

Categories

Questions about Iterative Development

What does Iterative Development do?

TDD iteration loops using Claude Code Stop hooks - runs tests after each response, feeds failures back automatically. Iterative Development is an agent skill from alinaqi/maggy.

When should I use Iterative Development?

Iterative Development fits situations like: tasks that involve Test-driven development.

How do I install Iterative Development in Claude Code?

Run `npx skills add alinaqi/maggy --skill iterative-development -a claude-code`. Or copy the skill folder (skills/iterative-development in alinaqi/maggy) into .claude/skills/iterative-development in your project. Claude Code loads it when a task matches its description.

How do I install Iterative Development in Codex?

Run `npx skills add alinaqi/maggy --skill iterative-development -a codex`. Or copy the skill folder (skills/iterative-development in alinaqi/maggy) into .agents/skills/iterative-development in your project. Codex loads it when a task matches its description.

Can I use Iterative Development in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add alinaqi/maggy --skill iterative-development -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/iterative-development, .gemini/skills/iterative-development, .github/skills/iterative-development and .opencode/skills/iterative-development in your project.

What does Iterative Development need to run?

Going by SKILL.md and its folder, Iterative Development needs the command-line tools its instructions call (npm, npx, pytest, ruff and mypy). Our summary lists: Python 3; Node.js.

Does Iterative Development access the network?

SKILL.md contains no URLs. Its commands use npm and npx, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Iterative Development safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Iterative Development use?

Iterative Development is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Iterative Development use?

About 2.2k tokens (SKILL.md is roughly 8.8k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Iterative Development?

Skills that share tags, products or a category with Iterative Development: TDD (pietheinstrengholt/rssmonster, 564 stars), TDD Workflow (hellangleZ/burn-in-cceverywhere-ralph, 112 stars), TDD (sanity-io/sanity, 6.4k stars) and Test Driven Development (farm-fe/farm, 5.6k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Iterative Development?

alinaqi (a GitHub user) maintains it in alinaqi/maggy, which has 707 GitHub stars. The repository holds 71 skills in this directory. The repository was last updated on September 24, 2026.

Source: alinaqi/maggy on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.