Fm/stg 474 more tests

.env.example → .env.example

@@ -1,4 +1,5 @@
-MODEL_API_KEY = "anthropic-or-openai-api-key"
+MODEL_API_KEY = "your-favorite-llm-api-key"
 BROWSERBASE_API_KEY = "browserbase-api-key"
 BROWSERBASE_PROJECT_ID = "browserbase-project-id"
 STAGEHAND_API_URL = "api_url"
+STAGEHAND_ENV= "LOCAL or BROWSERBASE"

.github/pull_request_template

@@ -3,3 +3,20 @@
 # what changed
 
 # test plan
+
+---
+
+## 🧪 Test Execution
+
+By default, **unit tests**, **integration tests**, and **smoke tests** run on all PRs.
+
+For additional testing, add one or more of these labels to your PR:
+
+- `test-browserbase` - Run Browserbase integration tests (requires API credentials)
+- `test-performance` - Run performance and load tests  
+- `test-llm` - Run LLM integration tests (requires API keys)
+- `test-e2e` - Run end-to-end workflow tests
+- `test-slow` - Run all slow-marked tests
+- `test-all` - Run the complete test suite (use sparingly)
+
+**Note**: Label-triggered tests only run when the labels are applied to the PR, not on individual commits.