/fusion-gemini

description: Fusion panel of Opus 4.8 + Gemini 3.1 Pro in parallel, judged by Opus 4.8 (opus4.8-gemini3.1pro)
argument-hint: <your question>

Invoke the **fusion** skill on the task below, forcing the `opus4.8-gemini3.1pro` panel: Opus 4.8 (Agent subagent) and Gemini 3.1 Pro (via `agy`, pseudo-TTY) answer the SAME prompt IN PARALLEL, each independently with web + bash and neither seeing the other's work → Opus 4.8 judges both answers → Opus writes the final answer grounded in the analysis.

Follow the skill's SKILL.md exactly (preflight → fan out in parallel → judge picking the track that fits the task → grounded final deliverable → save provenance → present). For a research/analysis task present the standard sections (Consensus / Contradictions / Partial coverage / Unique insights / Blind spots / Final answer); for a code/artifact task run both candidates and merge them into one working result with a merge rationale. Use exactly one Opus 4.8 panelist and one Gemini 3.1 Pro panelist — do NOT add a GPT-5.5 panelist or a second Opus run. Pass the task verbatim to both; no "lenses".

If the `agy` CLI is missing or the Gemini panelist fails/times out, drop it, record a one-line degradation note, and fall back to `opus4.8-4.8` (spawn a second independent Opus panelist) so the judge still sees two blind answers — never abort because one CLI failed.

Task: $ARGUMENTS