greenrun-cli 0.3.0 → 0.3.1

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "greenrun-cli",
-  "version": "0.3.0",
+  "version": "0.3.1",
   "description": "CLI and MCP server for Greenrun - browser test management for Claude Code",
   "type": "module",
   "main": "dist/server.js",

package/templates/claude-md.md

@@ -26,8 +26,14 @@ To run tests manually:
 
 1. Use `list_projects` to find the project
 2. Call `prepare_test_batch` with the project ID (and optional filter) to get test details and run IDs
-3. Execute each test's instructions using Playwright browser automation tools (`browser_navigate`, `browser_snapshot`, `browser_click`, `browser_type`)
-4. Call `complete_run` with the run ID, status (passed/failed/error), and a result summary
+3. For each test, launch a **Task agent** (subagent_type: "general-purpose", model: "haiku") that:
+   - Reads the test instructions via `export_test_instructions`
+   - Executes the test using Playwright browser automation tools
+   - Calls `complete_run` with the run ID, status, and result summary
+   - Returns a one-line summary
+4. Wait for each agent to complete before launching the next
+
+**Why Task agents?** Each test generates significant Playwright snapshot data. Running tests in Task agents keeps this data out of the parent context, preventing context bloat across a test suite.
 
 ### Auth Configuration
 
@@ -60,10 +66,15 @@ Don't ask the user for information you can derive from the codebase (base URL, l
 
 ### Creating Tests
 
-1. Navigate to the page you want to test using Playwright
-2. Write clear, step-by-step test instructions describing what to do and what to verify
-3. Use `create_page` to register the page URL if not already registered
-4. Use `create_test` with the instructions and page IDs
+Use a **Task agent** per test to keep browser snapshot data out of the parent context:
+
+1. Use `create_page` to register the page URL if not already registered
+2. For each test, launch a Task agent (subagent_type: "general-purpose", model: "haiku") that:
+   - Navigates to the page using Playwright
+   - Explores the UI and writes clear, step-by-step test instructions
+   - Calls `create_test` with the instructions and page IDs
+   - Returns a one-line summary
+3. After each agent completes, proceed to the next test
 
 ### Bug Detection During Test Creation
 

package/templates/commands/procedures.md

@@ -53,7 +53,7 @@ For each **unscripted** test:
 Task tool with:
 - subagent_type: "general-purpose"
 - max_turns: 30
-- model: "sonnet"
+- model: "haiku"
 - prompt: (see agent prompt below)
 ```
 
@@ -182,7 +182,7 @@ For each failed test:
 Task tool with:
 - subagent_type: "general-purpose"
 - max_turns: 25
-- model: "sonnet"
+- model: "haiku"
 - prompt: (see agent prompt below)
 ```