Ask both Codex (gpt-5.2) and Gemini a question. Use 'medium' or 'xhigh' for different reasoning levels.
Queries both Codex and Gemini in parallel, then synthesizes their responses into a unified analysis.
/plugin marketplace add kanlanc/Claude-Code-Kanlanc-Marketplace/plugin install ai-council@kanlanc-marketplaceQuery both Codex and Gemini in parallel, then synthesize their responses.
/council-chat [medium|xhigh] <question>
Parse the user's question and reasoning level
Run BOTH commands in parallel:
Codex:
codex exec --sandbox read-only -c model_reasoning_effort="<level>" "<question>"
Gemini:
gemini -s -y -o json "<question>"
Collect both responses
Synthesize and present the combined response
## AI Council Response
### Codex (GPT-5.2) Says:
[Codex response]
### Gemini Says:
[Gemini response]
### Council Synthesis:
**Agreement:** [Points both models agree on]
**Divergence:** [Where they differ]
**Recommendation:** [Synthesized recommendation]
---
*Session IDs: Codex=[id], Gemini=[id]*
*Use `/council-resume codex` or `/council-resume gemini` to continue*
/council-chat What's the best way to handle authentication?
/council-chat xhigh Design a caching strategy for this API
/council-chat medium Explain this function briefly