chore: import upstream snapshot with attribution

2026-07-13 13:25:13 +08:00
commit ec2b666284
2231 changed files with 491535 additions and 0 deletions
@@ -0,0 +1,46 @@
+# Output Schema with Tools Sample Agent
+
+This sample demonstrates how to use structured output (`output_schema`)
+alongside other tools in an ADK agent. Previously, this combination was not
+allowed, but now it's supported through a special processor that handles the
+interaction.
+
+## How it Works
+
+The agent combines:
+
+- **Tools**: `search_wikipedia` and `get_current_year` for gathering
+  information
+- **Structured Output**: `PersonInfo` schema to ensure consistent response
+  format
+
+When both `output_schema` and `tools` are specified:
+
+1. ADK automatically adds a special `set_model_response` tool
+1. The model can use the regular tools for information gathering
+1. For the final response, the model uses `set_model_response` with structured
+   data
+1. ADK extracts and validates the structured response
+
+## Expected Response Format
+
+The agent will return information in this structured format for user query
+
+> Tell me about Albert Einstein.
+
+```json
+{
+  "name": "Albert Einstein",
+  "age": 76,
+  "occupation": "Theoretical Physicist",
+  "location": "Princeton, New Jersey, USA",
+  "biography": "German-born theoretical physicist who developed the theory of relativity..."
+}
+```
+
+## Key Features Demonstrated
+
+1. **Tool Usage**: Agent can search Wikipedia and get current year
+1. **Structured Output**: Response follows strict PersonInfo schema
+1. **Validation**: ADK validates the response matches the schema
+1. **Flexibility**: Works with any combination of tools and output schemas