@bugzy-ai/bugzy 1.9.3 → 1.9.4

package/templates/init/CLAUDE.md

@@ -1,157 +1,157 @@
-# Project Memory
-
-## Project Overview
-**Type**: Test Management System
-**Purpose**: Autonomous QA testing with AI-based test generation and execution
-
-## SECURITY NOTICE
-
-**CRITICAL: Never Read the .env File**
-
-All agents, subagents, and slash commands must follow this security policy:
-
-- **NEVER read the `.env` file** - It contains secrets, credentials, and sensitive data
-- **ALWAYS use `.env.testdata`** instead - It contains only variable names, never actual values
-- **Reference variables by name only** - e.g., TEST_BASE_URL, TEST_USER_EMAIL (never read the actual values)
-- **This is enforced** - `.claude/settings.json` has `Read(.env)` in the deny list
-
-When you need to know what environment variables are available:
-1. Read `.env.testdata` to see variable names and structure
-2. Reference variables by name in documentation and instructions
-3. Trust that users will configure their own `.env` file from the example
-4. Never attempt to access actual credential values
-
-This policy applies to all agents, commands, and any code execution.
-
-## Agent System
-
-### Active Agents
-<!-- AUTO-GENERATED: This section is populated during project creation -->
-<!-- Configured agents will be listed here based on project setup -->
-
-Agents configured for this project are stored in `.claude/agents/`. These are specialized sub-agents that handle specific domains. The active agents and their integrations will be configured during project setup.
-
-### Agent Memory Index
-<!-- AUTO-GENERATED: Agent memory files are created during project setup -->
-
-Specialized memory files for domain experts will be created in `.bugzy/runtime/memory/` based on configured agents.
-
-## Project Structure
-
-### Core Components
-- **Test Plan**: `test-plan.md` - Generated via `/generate-test-plan` command using template
-- **Test Cases**: `./test-cases/` - Individual test case files generated via `/generate-test-cases`
-- **Exploration Reports**: `./exploration-reports/` - Structured reports from application exploration sessions
-- **Knowledge Base**: `.bugzy/runtime/knowledge-base.md` - Curated knowledge about the project (maintained using `.bugzy/runtime/knowledge-maintenance-guide.md`)
-- **Issue Tracking**: Managed by issue-tracker agent in your configured project management system
-- **Runtime**: `.bugzy/runtime/` - Project-specific runtime files:
-  - `memory/` - Agent memory and knowledge base files
-  - `templates/` - Document templates for test plan generation
-  - `knowledge-base.md` - Accumulated insights
-  - `project-context.md` - **CRITICAL**: Project SDLC, team information, QA workflow, and testing guidelines
-- **MCP Configuration**: `.bugzy/.mcp.json` - Template with all available MCP server configurations
-
-### Available Commands
-1. `/generate-test-plan` - Generate comprehensive test plan from product documentation using the documentation-researcher agent
-2. `/explore-application` - Systematically explore the application to discover UI elements, workflows, and behaviors, updating test plan with findings
-3. `/generate-test-cases` - Create E2E browser test cases (exploratory, functional, regression, smoke) based on documentation and test plan
-4. `/run-tests` - Execute test cases using the test-runner agent
-5. `/handle-message` - Process team responses and manage ongoing conversations with the product team using the team-communicator agent
-
-### Git Workflow
-
-**Git operations are now handled automatically by the execution environment.**
-
-Agents and commands should NOT perform git operations (commit, push). Instead:
-
-1. **Focus on core work**: Agents generate files and execute tests
-2. **External service handles git**: After task completion, the Cloud Run environment:
-   - Commits all changes with standardized messages
-   - Pushes to the remote repository
-   - Handles authentication and errors
-
-**Files Automatically Committed**:
-- `test-plan.md` - Test planning documents
-- `.env.testdata` - Environment variable templates
-- `./test-cases/*.md` - Individual test case files
-- `./exploration-reports/*.md` - Application exploration findings
-- `.bugzy/runtime/knowledge-base.md` - Accumulated testing insights
-- `.bugzy/runtime/memory/*.md` - Agent memory and knowledge base
-- `CLAUDE.md` - Updated project context (when modified)
-- `.bugzy/runtime/project-context.md` - Project-specific runtime context
-
-**Git-Ignored Files** (NOT committed):
-- `.env` - Local environment variables with secrets
-- `logs/` - Temporary log files
-- `tmp/` - Temporary files
-- `.playwright-mcp/` - Playwright videos (uploaded to GCS separately)
-- `node_modules/` - Node.js dependencies
-
-### Workflow
-
-#### Testing Workflow
-1. Generate test plan: `/generate-test-plan` command leverages the documentation-researcher agent to gather requirements
-2. Explore application: `/explore-application --focus [area] --depth [shallow|deep]` discovers actual UI elements and behaviors
-3. Create test cases: `/generate-test-cases --type [type] --focus [feature]` uses discovered documentation and exploration findings
-4. Execute tests: `/run-tests [test-id|type|tag|all]` runs test cases via test-runner agent
-5. Continuous refinement through exploration and test execution
-
-### Testing Lifecycle Phases
-1. **Initial Test Plan Creation** - From product documentation
-2. **Application Exploration** - Systematic discovery via `/explore-application`
-3. **Test Case Generation** - Stored as individual files with actual UI elements
-4. **Test Execution** - Automated runs via `/run-tests`
-5. **Continuous Refinement** - Through exploration and test execution
-6. **Regression Testing** - Scheduled or triggered execution
-
-## Key Project Insights
-<!-- Critical discoveries and patterns that apply across all domains -->
-
-## Cross-Domain Knowledge
-
-### Project Context Usage
-The `.bugzy/runtime/project-context.md` file contains critical project information that MUST be referenced by:
-- All agents during initialization to understand project specifics
-- All commands before processing to align with project requirements
-- Any custom implementations to maintain consistency
-
-Key information includes: QA workflow, story status management, bug reporting guidelines, testing environment details, and SDLC methodology.
-
-### Test Execution Standards
-
-**Automated Test Execution** (via `/run-tests` command):
-- Test runs are stored in hierarchical structure: `./test-runs/YYYYMMDD-HHMMSS/TC-XXX/exec-N/`
-- Each test run session creates:
-  - `execution-id.txt`: Contains BUGZY_EXECUTION_ID for session tracking
-  - `manifest.json`: Overall run metadata with all test cases and executions
-  - Per test case folders (`TC-001-login/`, etc.) containing execution attempts
-- Each execution attempt (`exec-1/`, `exec-2/`, `exec-3/`) contains:
-  - `result.json`: Playwright test result format with status, duration, errors, attachments
-  - `steps.json`: Step-by-step execution with video timestamps (if test uses `test.step()`)
-  - `video.webm`: Video recording (copied from Playwright's temp location)
-  - `trace.zip`: Trace file (only for failures)
-  - `screenshots/`: Screenshots directory (only for failures)
-- Videos are recorded for ALL tests, traces/screenshots only for failures
-- Custom Bugzy reporter handles all artifact organization automatically
-- External service uploads videos from execution folders to GCS
-
-**Step-Level Tracking** (Automated Tests):
-- Tests using `test.step()` API generate steps.json files with video timestamps
-- Each step includes timestamp and videoTimeSeconds for easy video navigation
-- High-level steps recommended (3-7 per test) for manageable navigation
-- Steps appear in both result.json (Playwright format) and steps.json (user-friendly format)
-- Tests without `test.step()` still work but won't have steps.json
-
-**Manual Test Execution** (via test-runner agent):
-- Manual test cases use separate format: `./test-runs/YYYYMMDD-HHMMSS/TC-XXX/`
-- Each test run generates:
-  - `summary.json`: Structured test result with video filename reference
-  - `steps.json`: Step-by-step execution with timestamps and video synchronization
-- Video recording via Playwright MCP --save-video flag
-- Videos remain in `.playwright-mcp/` folder - external service uploads to GCS
-
-**General Rules**:
-- NO test-related files should be created in project root
-- DO NOT copy, move, or delete video files manually
-- DO NOT CREATE ANY SUMMARY, REPORT, OR ADDITIONAL FILES unless explicitly requested
-<!-- Additional cross-domain information below -->
+# Project Memory
+
+## Project Overview
+**Type**: Test Management System
+**Purpose**: Autonomous QA testing with AI-based test generation and execution
+
+## SECURITY NOTICE
+
+**CRITICAL: Never Read the .env File**
+
+All agents, subagents, and slash commands must follow this security policy:
+
+- **NEVER read the `.env` file** - It contains secrets, credentials, and sensitive data
+- **ALWAYS use `.env.testdata`** instead - It contains only variable names, never actual values
+- **Reference variables by name only** - e.g., TEST_BASE_URL, TEST_USER_EMAIL (never read the actual values)
+- **This is enforced** - `.claude/settings.json` has `Read(.env)` in the deny list
+
+When you need to know what environment variables are available:
+1. Read `.env.testdata` to see variable names and structure
+2. Reference variables by name in documentation and instructions
+3. Trust that users will configure their own `.env` file from the example
+4. Never attempt to access actual credential values
+
+This policy applies to all agents, commands, and any code execution.
+
+## Agent System
+
+### Active Agents
+<!-- AUTO-GENERATED: This section is populated during project creation -->
+<!-- Configured agents will be listed here based on project setup -->
+
+Agents configured for this project are stored in `.claude/agents/`. These are specialized sub-agents that handle specific domains. The active agents and their integrations will be configured during project setup.
+
+### Agent Memory Index
+<!-- AUTO-GENERATED: Agent memory files are created during project setup -->
+
+Specialized memory files for domain experts will be created in `.bugzy/runtime/memory/` based on configured agents.
+
+## Project Structure
+
+### Core Components
+- **Test Plan**: `test-plan.md` - Generated via `/generate-test-plan` command using template
+- **Test Cases**: `./test-cases/` - Individual test case files generated via `/generate-test-cases`
+- **Exploration Reports**: `./exploration-reports/` - Structured reports from application exploration sessions
+- **Knowledge Base**: `.bugzy/runtime/knowledge-base.md` - Curated knowledge about the project (maintained using `.bugzy/runtime/knowledge-maintenance-guide.md`)
+- **Issue Tracking**: Managed by issue-tracker agent in your configured project management system
+- **Runtime**: `.bugzy/runtime/` - Project-specific runtime files:
+  - `memory/` - Agent memory and knowledge base files
+  - `templates/` - Document templates for test plan generation
+  - `knowledge-base.md` - Accumulated insights
+  - `project-context.md` - **CRITICAL**: Project SDLC, team information, QA workflow, and testing guidelines
+- **MCP Configuration**: `.bugzy/.mcp.json` - Template with all available MCP server configurations
+
+### Available Commands
+1. `/generate-test-plan` - Generate comprehensive test plan from product documentation using the documentation-researcher agent
+2. `/explore-application` - Systematically explore the application to discover UI elements, workflows, and behaviors, updating test plan with findings
+3. `/generate-test-cases` - Create E2E browser test cases (exploratory, functional, regression, smoke) based on documentation and test plan
+4. `/run-tests` - Execute test cases using the test-runner agent
+5. `/handle-message` - Process team responses and manage ongoing conversations with the product team using the team-communicator agent
+
+### Git Workflow
+
+**Git operations are now handled automatically by the execution environment.**
+
+Agents and commands should NOT perform git operations (commit, push). Instead:
+
+1. **Focus on core work**: Agents generate files and execute tests
+2. **External service handles git**: After task completion, the Cloud Run environment:
+   - Commits all changes with standardized messages
+   - Pushes to the remote repository
+   - Handles authentication and errors
+
+**Files Automatically Committed**:
+- `test-plan.md` - Test planning documents
+- `.env.testdata` - Environment variable templates
+- `./test-cases/*.md` - Individual test case files
+- `./exploration-reports/*.md` - Application exploration findings
+- `.bugzy/runtime/knowledge-base.md` - Accumulated testing insights
+- `.bugzy/runtime/memory/*.md` - Agent memory and knowledge base
+- `CLAUDE.md` - Updated project context (when modified)
+- `.bugzy/runtime/project-context.md` - Project-specific runtime context
+
+**Git-Ignored Files** (NOT committed):
+- `.env` - Local environment variables with secrets
+- `logs/` - Temporary log files
+- `tmp/` - Temporary files
+- `.playwright-mcp/` - Playwright videos (uploaded to GCS separately)
+- `node_modules/` - Node.js dependencies
+
+### Workflow
+
+#### Testing Workflow
+1. Generate test plan: `/generate-test-plan` command leverages the documentation-researcher agent to gather requirements
+2. Explore application: `/explore-application --focus [area] --depth [shallow|deep]` discovers actual UI elements and behaviors
+3. Create test cases: `/generate-test-cases --type [type] --focus [feature]` uses discovered documentation and exploration findings
+4. Execute tests: `/run-tests [test-id|type|tag|all]` runs test cases via test-runner agent
+5. Continuous refinement through exploration and test execution
+
+### Testing Lifecycle Phases
+1. **Initial Test Plan Creation** - From product documentation
+2. **Application Exploration** - Systematic discovery via `/explore-application`
+3. **Test Case Generation** - Stored as individual files with actual UI elements
+4. **Test Execution** - Automated runs via `/run-tests`
+5. **Continuous Refinement** - Through exploration and test execution
+6. **Regression Testing** - Scheduled or triggered execution
+
+## Key Project Insights
+<!-- Critical discoveries and patterns that apply across all domains -->
+
+## Cross-Domain Knowledge
+
+### Project Context Usage
+The `.bugzy/runtime/project-context.md` file contains critical project information that MUST be referenced by:
+- All agents during initialization to understand project specifics
+- All commands before processing to align with project requirements
+- Any custom implementations to maintain consistency
+
+Key information includes: QA workflow, story status management, bug reporting guidelines, testing environment details, and SDLC methodology.
+
+### Test Execution Standards
+
+**Automated Test Execution** (via `/run-tests` command):
+- Test runs are stored in hierarchical structure: `./test-runs/YYYYMMDD-HHMMSS/TC-XXX/exec-N/`
+- Each test run session creates:
+  - `execution-id.txt`: Contains BUGZY_EXECUTION_ID for session tracking
+  - `manifest.json`: Overall run metadata with all test cases and executions
+  - Per test case folders (`TC-001-login/`, etc.) containing execution attempts
+- Each execution attempt (`exec-1/`, `exec-2/`, `exec-3/`) contains:
+  - `result.json`: Playwright test result format with status, duration, errors, attachments
+  - `steps.json`: Step-by-step execution with video timestamps (if test uses `test.step()`)
+  - `video.webm`: Video recording (copied from Playwright's temp location)
+  - `trace.zip`: Trace file (only for failures)
+  - `screenshots/`: Screenshots directory (only for failures)
+- Videos are recorded for ALL tests, traces/screenshots only for failures
+- Custom Bugzy reporter handles all artifact organization automatically
+- External service uploads videos from execution folders to GCS
+
+**Step-Level Tracking** (Automated Tests):
+- Tests using `test.step()` API generate steps.json files with video timestamps
+- Each step includes timestamp and videoTimeSeconds for easy video navigation
+- High-level steps recommended (3-7 per test) for manageable navigation
+- Steps appear in both result.json (Playwright format) and steps.json (user-friendly format)
+- Tests without `test.step()` still work but won't have steps.json
+
+**Manual Test Execution** (via test-runner agent):
+- Manual test cases use separate format: `./test-runs/YYYYMMDD-HHMMSS/TC-XXX/`
+- Each test run generates:
+  - `summary.json`: Structured test result with video filename reference
+  - `steps.json`: Step-by-step execution with timestamps and video synchronization
+- Video recording via Playwright MCP --save-video flag
+- Videos remain in `.playwright-mcp/` folder - external service uploads to GCS
+
+**General Rules**:
+- NO test-related files should be created in project root
+- DO NOT copy, move, or delete video files manually
+- DO NOT CREATE ANY SUMMARY, REPORT, OR ADDITIONAL FILES unless explicitly requested
+<!-- Additional cross-domain information below -->

package/templates/init/test-runs/README.md

@@ -1,45 +1,45 @@
-# Test Runs
-
-This directory contains execution results and artifacts from test runs.
-
-## Structure
-
-Test runs are organized by timestamp and test case:
-```
-YYYYMMDD-HHMMSS/
-├── TC-XXX/
-│   ├── summary.json
-│   ├── steps.json
-│   └── video.webm
-```
-
-## Contents
-
-Each test run includes:
-- **summary.json**: Structured test result with video metadata and status
-- **steps.json**: Detailed execution steps with timestamps and video synchronization
-- **video.webm**: Video recording of the entire test execution
-
-**Note**: All test information (status, failures, step details, observations) is captured in the structured JSON files. The video provides visual evidence synchronized with the steps.
-
-## Test Result Schema
-
-All test results follow the schema defined in `.bugzy/runtime/templates/test-result-schema.md`.
-
-Key features:
-- Video recording with synchronized step navigation
-- Timestamped steps for precise playback control
-- Structured metadata for test status, type, and priority
-- User-friendly action descriptions for easy understanding
-
-## Usage
-
-Test runs are automatically generated when:
-- Tests are executed via automation
-- Manual test execution is logged
-- Event processing triggers test validation
-
-Results are used for:
-- Issue tracking via the issue-tracker agent
-- Learning extraction for continuous improvement
-- Regression analysis and pattern recognition
+# Test Runs
+
+This directory contains execution results and artifacts from test runs.
+
+## Structure
+
+Test runs are organized by timestamp and test case:
+```
+YYYYMMDD-HHMMSS/
+├── TC-XXX/
+│   ├── summary.json
+│   ├── steps.json
+│   └── video.webm
+```
+
+## Contents
+
+Each test run includes:
+- **summary.json**: Structured test result with video metadata and status
+- **steps.json**: Detailed execution steps with timestamps and video synchronization
+- **video.webm**: Video recording of the entire test execution
+
+**Note**: All test information (status, failures, step details, observations) is captured in the structured JSON files. The video provides visual evidence synchronized with the steps.
+
+## Test Result Schema
+
+All test results follow the schema defined in `.bugzy/runtime/templates/test-result-schema.md`.
+
+Key features:
+- Video recording with synchronized step navigation
+- Timestamped steps for precise playback control
+- Structured metadata for test status, type, and priority
+- User-friendly action descriptions for easy understanding
+
+## Usage
+
+Test runs are automatically generated when:
+- Tests are executed via automation
+- Manual test execution is logged
+- Event processing triggers test validation
+
+Results are used for:
+- Issue tracking via the issue-tracker agent
+- Learning extraction for continuous improvement
+- Regression analysis and pattern recognition