Select the OpenCode adapter from canonical config in ai_check/ai_test;
add --ai-backend/--opencode-* flags to the installer and sys. Require a
validated interaction (not a provider name) for the mandatory-AI gate in
both the bridge and the CLI, invalidated by config, model or credential
changes. Tests cover the gate, invalidation and the OpenCode backend.
Installing Phoebus now requires a connected model: the install gate refuses
with no provider unless the explicit break-glass --no-ai (alias --legacy, or
PHOEBUS_NO_AI=1) is used. Add a 'local' provider kind (Ollama/llama.cpp/
LM Studio) with an optional API key and --ai-endpoint override, so a user can
run a model they downloaded; no model is shipped. Desktop installs now include
xdg-desktop-portal + gtk/wlr backends for Flatpak and Wayland apps.
Add an OpenAI-compatible provider layer (DeepSeek, OpenAI and a test
transport) that sends only bounded system facts and injects the API key via a
curl config on stdin, never in argv, config or logs. Extend the typed proposal
schema with scoped package.use, build jobs and binary-package preference; every
change now needs a non-empty reason and duplicates are rejected. Wire
--ai/--ai-proposal-file/--ai-facts into the CLI and add provider selection to
the TUI. Deterministic validation and review are unchanged: the model never
reaches the executor and the no-AI path keeps working.