The PYRANA_PLATFORM example supports its claim with quotes stating that PyranaHarness implements native tool calling via LiteLLM, creates an initial Turn 0 plan, repeats a native tool-calling loop, and finalizes by building and persisting an AgentOutputEnvelope.

supporting_quotes:

— PYRANA_Internal_Knowledge_Graph_Specification.md

"PyranaHarness (pyrana_services/orchestration/harness/pyrana_harness.py)

— PYRANA_Internal_Knowledge_Graph_Specification.md

which implements native tool calling via LiteLLM"

— PYRANA_Internal_Knowledge_Graph_Specification.md

"Turn 0: create initial plan via LLM"

— PYRANA_Internal_Knowledge_Graph_Specification.md

"Native tool calling loop: LLM emits tool_calls → execute tools → observe → repeat"

— PYRANA_Internal_Knowledge_Graph_Specification.md

"Finalize: build and persist AgentOutputEnvelope"

— PYRANA_Internal_Knowledge_Graph_Specification.md