Implement end-to-end hypothesis generation and testing workflow for issue #14

[taivu1998]

Summary

This draft PR implements the end-to-end hypothesis workflow for issue #14 on top of the current main branch snapshot.

It adds a shared hypothesis engine that supports both hypothesis generation and hypothesis testing, wires that engine into the application and orchestration layers, and adds the supporting configs, scripts, and focused tests needed to exercise the flow.

What Changed

• add shared hypothesis datatypes, prompts, agents, tools, and a dedicated hypothesis workflow state machine
• add a dedicated DeepResearch.hypothesis_app entrypoint and CLI wiring for hypothesis runs
• route explicit hypothesis generation and testing flows through DeepResearch.app and the primary workflow orchestrator
• add hypothesis config groups, workflow compatibility shims, and runnable example configs
• harden adjacent integration surfaces uncovered during implementation, including optional tool input validation and lazy tool loading
• add focused import, workflow, tool, and example tests, plus a smoke script for the new workflow

Why

Issue #14 calls for comprehensive hypothesis testing support. On current main, the hypothesis path was still effectively placeholder-backed: the workflow was not fully wired end to end, testing-environment generation was incomplete, and several integration edges around provided hypotheses, routing, and validation still needed real implementation.

This PR closes that gap with a single shared workflow that keeps the implementation cohesive without adding unnecessary orchestration layers.

Validation

• UV_CACHE_DIR=/tmp/uv-cache uv run pytest tests/test_statemachines/test_hypothesis_workflow.py tests/test_tools/test_hypothesis_tools.py tests/test_examples/test_hypothesis_engine.py tests/imports/test_app_imports.py tests/imports/test_datatypes_imports.py tests/imports/test_tools_imports.py -q
• UV_CACHE_DIR=/tmp/uv-cache uv run python scripts/test/run_hypothesis_smoke.py
• UV_CACHE_DIR=/tmp/uv-cache uv run pytest -q

Issue

Addresses #14.

[codecov]

Codecov Report

❌ Patch coverage is 95.74803% with 27 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
...pResearch/src/statemachines/hypothesis_workflow.py	92.18%	15 Missing ⚠️
DeepResearch/src/tools/hypothesis_tools.py	97.43%	6 Missing ⚠️
DeepResearch/src/agents/workflow_orchestrator.py	88.23%	2 Missing ⚠️
DeepResearch/src/datatypes/hypothesis.py	96.61%	2 Missing ⚠️
DeepResearch/app.py	97.56%	1 Missing ⚠️
DeepResearch/hypothesis_app.py	92.85%	1 Missing ⚠️

📢 Thoughts on this report? Let us know!

[greptile-apps]

Your free trial has ended. If you'd like to continue receiving code reviews, you can add a payment method here.

[Josephrp]

@taivu1998 please retarget the PR from "to main" to "to dev" please :-) otherwise it's 10/10 , with all my thanks

[Josephrp]

LGTM 🏆

The base branch was changed.

[greptile-apps]

Your free trial has ended. If you'd like to continue receiving code reviews, you can add a payment method here.

[taivu1998]

Thanks @Josephrp, I just updated. Could you review and merge?