Python: updated declarative samples and handling of non-pydantic response formats

[Eduard van Valkenburg (eavanvalkenburg)]

Motivation and Context

Fixed the declarative samples and noticed that non-pydantic response formats (json schema definition) was not handled in the response types.
Fixed now across the board.

Description

Contribution Checklist

• The code builds clean without any errors or warnings
• The PR follows the Contribution Guidelines
• All unit tests pass, and I have added new tests where possible
• Is this a breaking change? If yes, add "[BREAKING]" prefix to the title of the PR.

[Mark Wallace (markwallace-microsoft)]

Python Test Coverage Report •

File	Stmts	Miss	Cover	Missing
packages/core/agent_framework
_agents.py	412	52	87%	461, 470, 525, 1020, 1065, 1138–1142, 1202, 1230, 1267, 1288, 1308–1309, 1314, 1361, 1403, 1425, 1427, 1440, 1446, 1491, 1493, 1502–1507, 1512, 1514, 1520–1521, 1528, 1530–1531, 1539–1540, 1543–1545, 1555–1560, 1564, 1569, 1571
_clients.py	137	7	94%	324, 375, 531–534, 649
_tools.py	947	88	90%	190–191, 364, 366, 379, 404–406, 414, 432, 446, 453, 460, 483, 485, 492, 500, 539, 583, 587, 619–621, 623, 629, 674–676, 678, 701, 727, 731, 769–771, 775, 797, 909–915, 951, 963, 965, 967, 970–973, 994, 998, 1002, 1016–1018, 1359, 1381, 1468–1474, 1603, 1607, 1653, 1714–1715, 1830, 1850, 1852, 1908, 1971, 2143–2144, 2164, 2220–2221, 2281, 2359–2360, 2427, 2432, 2439
_types.py	1075	90	91%	58, 67–68, 122, 127, 146, 148, 152, 156, 158, 160, 162, 180, 184, 210, 232, 237, 242, 246, 276, 687–688, 847–848, 1233, 1305, 1340, 1360, 1370, 1422, 1554–1556, 1746, 1837–1842, 1867, 1955, 1961–1963, 1968, 2059, 2071, 2094, 2349, 2373, 2468, 2717, 2923, 2992, 3003, 3005–3009, 3011, 3014–3022, 3032, 3236–3238, 3241–3243, 3247, 3252, 3256, 3340–3342, 3371, 3448–3452
packages/declarative/agent_framework_declarative
_loader.py	253	58	77%	574, 576, 603–613, 655–658, 661–663, 666, 668–670, 673, 681, 698–703, 755–760, 778–783, 791, 793, 798, 801–804, 807–808, 811–812, 826–827, 857
packages/ollama/agent_framework_ollama
_chat_client.py	208	9	95%	396, 398, 409, 413–414, 422, 433, 512, 518
packages/openai/agent_framework_openai
_chat_client.py	855	127	85%	520–523, 527–528, 533–534, 544–545, 552, 567–573, 594, 602, 625, 743, 842, 901, 903, 905, 907, 973, 987, 1067, 1077, 1082, 1125, 1202, 1232, 1289, 1382, 1387, 1391–1393, 1397–1398, 1464, 1493, 1499, 1509, 1515, 1520, 1526, 1531–1532, 1593, 1615–1616, 1631–1632, 1650–1651, 1692–1695, 1857, 1895–1896, 1912, 1914, 1992–2000, 2122, 2157, 2172, 2192–2202, 2215, 2226–2230, 2244, 2258–2269, 2278, 2310–2313, 2321–2322, 2324–2326, 2340–2342, 2352–2353, 2359, 2374
TOTAL	26994	3187	88%

Python Unit Test Overview

Tests	Skipped	Failures	Errors	Time
5341	20 💤	0 ❌	0 🔥	1m 28s ⏱️

[Copilot]

Pull request overview

This PR updates Python declarative samples and extends structured output handling so response_format values provided as runtime JSON schema mappings (dicts) are preserved and parsed into response.value across multiple chat clients.

Changes:

• Update declarative samples/YAML to align with Foundry environment variables and loader defaults.
• Propagate non-Pydantic response_format mappings through core response types so JSON text is parsed into dict values.
• Add/adjust tests across providers (OpenAI, Azure OpenAI, Foundry, Anthropic, Ollama, core) to cover dict response_format parsing.

Reviewed changes

Copilot reviewed 24 out of 24 changed files in this pull request and generated 5 comments.

Show a summary per file

File	Description
python/samples/05-end-to-end/evaluation/foundry_evals/evaluate_tool_calls_sample.py	Minor sample instruction string formatting.
python/samples/04-hosting/a2a/a2a_agent_as_function_tools.py	Minor formatting of env var error message.
python/samples/02-agents/declarative/openai_agent.py	Use create_agent_from_yaml_path instead of manual YAML loading.
python/samples/02-agents/declarative/microsoft_learn_agent.py	Add explanatory module-level documentation for the sample.
python/samples/02-agents/declarative/inline_yaml.py	Update sample env var guidance and inline YAML to Foundry model variable.
python/packages/openai/tests/openai/test_openai_chat_completion_client.py	Add test coverage for dict response_format parsing; update integration expectations.
python/packages/openai/tests/openai/test_openai_chat_client.py	Add runtime JSON schema mapping parsing-path test; update integration expectations.
python/packages/openai/tests/openai/test_openai_chat_client_azure.py	Update integration expectations for runtime JSON schema parsing.
python/packages/openai/agent_framework_openai/_chat_client.py	Preserve non-Pydantic response_format on ChatResponse for parsing.
python/packages/ollama/tests/test_ollama_chat_client.py	Add test for dict response_format parsing.
python/packages/ollama/agent_framework_ollama/_chat_client.py	Thread response_format through Ollama parsing to enable JSON parsing.
python/packages/foundry/tests/foundry/test_foundry_chat_client.py	Add parsing-path test; update integration expectations.
python/packages/declarative/agent_framework_declarative/_loader.py	Update provider mapping and change default_provider to Foundry.
python/packages/core/tests/core/test_types.py	Add core type tests for mapping response_format parsing (sync + streaming).
python/packages/core/tests/core/test_observability.py	Ensure streaming finalizer passes through mapping response_format.
python/packages/core/tests/core/test_agents.py	Add agent-level tests for mapping response_format propagation/parsing.
python/packages/core/tests/core/conftest.py	Ensure test streaming clients pass through mapping response_format.
python/packages/core/agent_framework/_types.py	Add mapping structured format support and JSON parsing for value.
python/packages/core/agent_framework/_tools.py	Pass mapping response_format through streaming finalizer path.
python/packages/core/agent_framework/_evaluation.py	Minor formatting tweaks.
python/packages/core/agent_framework/_clients.py	Pass mapping response_format into ChatResponse.from_updates.
python/packages/core/agent_framework/_agents.py	Preserve mapping response_format when constructing AgentResponse.
python/packages/anthropic/tests/test_anthropic_client.py	Add test ensuring dict response_format is preserved for parsing.
agent-samples/foundry/MicrosoftLearnAgent.yaml	Update env var names to FOUNDRY_* conventions.