Repository navigation
Post-wizard Ollama validation before session start #53
Description
Problem Statement
After #43, users get early Ollama checks for YAML defaults only. The interactive wizard lets users invite custom agents and set orchestrator models interactively. Those models are not validated before run_session() starts, so users still hit LiteLLM/Ollama errors mid-session after passing startup preflight (especially with --skip-preflight or non-Ollama defaults + Ollama custom agents).
Proposed Solution
-
After wizard builds
active_agent_configs(+ optional orchestrator), collect allollama/…models from the final session config. -
Reuse shared Ollama helper (issue feat: add timeout to Agent.generate_response() #2) to verify tags once before
run_session(). -
On failure: print the same actionable panel as preflight; offer to return to wizard, edit model strings, or abort (do not start session with known-bad tags).
-
Respect
--skip-preflightfor both startup and post-wizard checks (or document--skip-preflightas global bypass). -
Smoke test: mock validation hook; assert called with wizard-built configs.
Acceptance criteria
- Custom agent with missing Ollama tag blocked before first turn
- Orchestrator chosen in wizard validated even when YAML default is non-Ollama
-
--skip-preflightbypasses both checks - No duplicate
/api/tagscalls when tags unchanged from startup check
Alternatives Considered
- Only fix at inference (feat: surface inference failures in CLI with recovery hints (not as agent speech) #37) — better UX but wastes wizard setup time.
- Move all preflight to post-wizard only — rejected; YAML-only users lose early exit.
Priority
Medium 🟡
Additional Context
No response
Activity