…, framework-only can_handle)

Mirror radu's LangChain-adapter review across the Google ADK adapter:
- __init__: drop the import-time registration side-effect; registration only via the uipath.governance.adapters entry point.
- can_handle: claim only a real google.adk BaseAgent; remove the duck-typed (name + model-callback / sub_agents) fallback.
- docstring: 'governance host' instead of uipath-runtime internals.
- tests: can_handle uses a real LlmAgent; duck-typed look-alikes are now rejected.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

…K adapter

Record that LlmRequest/LlmResponse/content/parts are read via getattr
rather than isinstance on ADK's typed models, to avoid hard-coupling to
google-adk internals and to let tests duck-type payloads without the
package installed.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

Core PR #1761 removed BaseAdapter from uipath-core. Migrate to the
factory-evaluator pattern (matching #899):

- governance/adapter.py -> callbacks.py: replace the BaseAdapter subclass
  (name/can_handle/attach/detach) with module-level install_governance()
  that installs governance on each LlmAgent's native *_callback slots
  (walking sub_agents); keep GovernanceCallbacks + the callback helpers;
  drop the detach-only _remove_callbacks. File named for its seam
  (callbacks), like LangChain's callbacks.py.
- runtime/factory.py: new_runtime reads `evaluator` from kwargs and
  calls install_governance before building the Runner.
- governance/__init__.py: drop register_governance_adapter + registry
  import; expose install_governance. No import-time side effects.
- pyproject.toml: remove the uipath.governance.adapters entry point.
- tests (test_adapter.py -> test_callbacks.py): drop can_handle/attach/
  detach; cover install_governance + factory wiring.

ruff + mypy clean; 21 governance tests pass.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

…ool walk

Review findings (Viswa) for PR #362:

- Drop the per-callbacks uuid trace_id (identical for every call); trace
  correlation is owned by the layer below, matching LangChain. Requires
  uipath-core >= 0.5.20, which removed trace_id from EvaluatorProtocol —
  bumped in the lock.
- Count llm/tool calls only after governance passes: a DENY raised before
  the bump, inflating the counter on blocked calls.
- Follow AgentTool-wrapped agents: an agent exposed as a tool lives in
  ``tools`` (on ``tool.agent``), not ``sub_agents``, so the walk missed it.
- Refresh governance metadata on cached-agent reuse: the factory caches
  agents by entrypoint, so a second new_runtime with a new session_id would
  otherwise keep the first run's session_id (install was a no-op). Added
  GovernanceCallbacks.rebind() + _find_governance_callbacks().
- Log (not silently drop) when the tree walk hits the node cap.
- Cap _stringify output so an oversized tool result / args / response
  can't hand a multi-megabyte string to the evaluator.
- Trailing newline (W292).

Tests: added AgentTool-walk, cached-agent rebind, and no-inflation-on-block
coverage.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

Follow-up to the #362 review pass — the earlier commit covered AgentTool walk,
_stringify cap, cached-agent rebind, trace_id drop, and the cap warning; these
are the rest of page 6:

- callbacks.py:249 (Major) — before_tool no longer passes tool args uncapped.
  New _cap_args bounds the payload: within budget the dict passes through
  unchanged (per-key rules still work); once its serialized size exceeds the
  cap it is replaced with a single capped {"_truncated": ...}. Contrast with
  after_tool, which already capped its result.
- test_callbacks.py (Minor) — the tree test now also asserts the container
  agent is NOT decorated; added a huge-args cap test. (Double-attach with a
  pre-existing user callback is already covered by
  test_install_governance_preserves_existing_callback_and_runs_first.)

Acknowledged, unchanged (flagged as follow-ups):
- callbacks sync-only (Major): making governance async is a protocol-wide
  change (the evaluator is sync) — deferred, not one-off here.
- _session_state accumulation across cached-agent reuse (Minor): fixed by the
  rebind added in the earlier commit (rebind resets the counters). Per-task
  safety for ParallelAgent's concurrent callbacks is a known limitation.
- LIFO cap ordering (Minor): the cap now logs; which nodes are dropped past
  the cap is non-deterministic but all found nodes are governed.

Tests: 66 pass.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

Same pre-existing CI-lint failure (mypy runs over tests):
- FakeEvaluator evaluate_* -> (self, *args, **kwargs) -> Any so it satisfies
  EvaluatorProtocol; bare dict -> dict[str, Any].
- moved the Boom test-double arg-type ignore onto the reported line;
  func-returns-value ignores on the return-None pass-through asserts.

mypy . clean (26 files); ruff + 66 tests green.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

New-code coverage ~89% -> over the 90% gate. Added non-block swallow on the
model/tool callbacks and _content_text / _cap_args / _stringify helper edges.
governance/callbacks.py: 89% -> 96%.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>