conductor(plan): write 6-phase implementation plan for qwen_llama_grok_integration_20260606

Browse Source

~30 tasks across 6 phases, each with explicit Red-Green-Refactor TDD steps:
- Phase 1 (1.1-1.8): Capability matrix framework (src/vendor_capabilities.py)
  + shared OpenAI-compatible helper (src/openai_compatible.py). 13 unit tests.
- Phase 2 (2.1-2.8): Qwen via DashScope native SDK. 5 unit tests.
- Phase 3 (3.1-3.7): Grok (xAI) + Llama (Ollama + OpenRouter + custom URL)
  via shared helper. 8 unit tests.
- Phase 4 (4.1-4.3): MiniMax refactor (_send_minimax from ~250 -> ~50 lines).
  Safety net: existing tests/test_minimax_provider.py.
- Phase 5 (5.1-5.5): 9 capability-driven UX adaptations in src/gui_2.py.
  Manual smoke test for all 3 new vendors.
- Phase 6 (6.1-6.4): Update docs/guide_ai_client.md + guide_models.md.
  Archive the track.

Data-oriented design: shared helper is the algorithm on normalized data;
_send_<vendor>() entry points are thin boundary adapters.

1-space indentation per project style guide. No placeholders. All test
code is concrete. Self-review at end confirms spec coverage (every
section of spec.md mapped to a task).

...

This commit is contained in:

Edward R. Gonzalez

2026-06-06 15:06:30 -04:00

parent 97daaff29b

commit b17cbbdeca

1 changed files with 2177 additions and 0 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download Patch File Download Diff File Expand all files Collapse all files

conductor/tracks/qwen_llama_grok_integration_20260606/plan.md

+2177

File diff suppressed because it is too large Load Diff