📋 QA Planner Agent

Reads your requirements.md and writes a professional test_plan.md — acting as a Senior QA Lead with 15+ years of experience.

Part of the @agents-forge AIQA pipeline:

analyst → qa-planner → qa-engineer → qa-reviewer → qa-reporter

Install

npm install @agents-forge/qa-planner

Quick Start

# Generate requirements first (if you haven't already)
npx @agents-forge/analyst https://my-app.com requirements.md

# Then generate your test plan
npx @agents-forge/qa-planner ./requirements.md test_plan.md

Authentication

No config needed — auto-detected in this order:

| Method | How to set up | |---|---| | Anthropic API key | Add ANTHROPIC_API_KEY=sk-ant-... to .env | | Claude subscription | npm install -g @anthropic-ai/claude-code → claude login | | GitHub Copilot | gh auth login |

Features

• 📖 Reads requirements.md and understands every feature and user flow
• 🧠 Acts as a Senior QA Lead — thinks in test coverage, not just checklists
• 🎯 Maps features to test types: smoke, regression, negative, edge cases
• ⚠️ Identifies risk areas and assigns P0/P1/P2 priority
• 🏷️ Defines a test tagging model ready for Playwright automation
• 🔐 Multi-provider auth — works with API key, Claude login, or GitHub Copilot
• ⚡ Streams output in real time so you can see it working

CLI Usage

# Basic — prompts for output filename
npx @agents-forge/qa-planner ./requirements.md

# With output filename set
npx @agents-forge/qa-planner ./requirements.md test_plan.md

Programmatic Usage

import { runQAPlanner } from "@agents-forge/qa-planner";

await runQAPlanner({
  requirementsFile: "./requirements.md",   // required
  outputFile: "test_plan.md",              // optional — prompted if omitted
  cwd: process.cwd(),                      // optional
});

Output Format

The agent writes a test_plan.md with these sections:

| Section | Content | |---|---| | Document Info | Version, date, status | | Executive Summary | What's being tested and why | | Scope | In scope / out of scope features | | Test Objectives | Specific, measurable quality goals | | Test Strategy | Approach for each test type | | Feature Areas | Every feature with test type + priority | | Risk Analysis | Risks with likelihood, impact, mitigation | | Test Environments | Browser matrix, test data needs | | Entry & Exit Criteria | When to start and stop testing | | Test Tagging Model | Tags for Playwright automation | | Assumptions | Dependencies and assumptions | | Sign-off | QA Lead approval section |

Environment Variables

| Variable | Required | Description | |---|---|---| | ANTHROPIC_API_KEY | Optional | Anthropic API key | | QA_PLANNER_MODEL | Optional | Override the model used |

Troubleshooting

| Error | Fix | |---|---| | Requirements file not found | Run @agents-forge/analyst first to generate requirements.md | | Authentication failed | Set ANTHROPIC_API_KEY or run claude login or gh auth login | | File not written | Check terminal output — agent may have hit an error mid-run | | claude: command not found | npm install -g @anthropic-ai/claude-code |

Full Pipeline

# 1. Analyse website → requirements.md
npx @agents-forge/analyst https://my-app.com requirements.md

# 2. Plan testing → test_plan.md
npx @agents-forge/qa-planner requirements.md test_plan.md

# 3. Write tests → test_cases.md + tests/*.spec.ts  (coming soon)
npx @agents-forge/qa-engineer test_plan.md

# 4. Review quality → review_report.md  (coming soon)
npx @agents-forge/qa-reviewer

# 5. Run & report → summary_report.md  (coming soon)
npx @agents-forge/qa-reporter

License

MIT