What it does
This setting uses Claude Opus 5.5 as an advisor. The main model consults it before committing to an approach, when it hits a recurring error, and before declaring a task done. It can advise Haiku, Sonnet and Opus up to 5.5, but not Fable. It is pinned to the full model ID, so the advisor stays on 5.5.
Someone would reach for it when they want a stronger model reviewing the main model's choices at those checkpoints, rather than relying on the main model alone.
It runs inside Claude Code. On the Anthropic API it needs Claude Code v2.1.280 or later; on Claude Platform on AWS it needs v2.1.300 or later. On the Anthropic API the advisor stays off when DISABLE_TELEMETRY or CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC is set.
Limitations and requirements
Requires Claude Code v2.1.280 or later on the Anthropic API, or v2.1.300 or later on Claude Platform on AWS. On the Anthropic API the advisor is disabled when DISABLE_TELEMETRY or CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC is set. Cannot advise the Fable model.
Safety
Our team has not reviewed this resource. Only automated rules checked it; resources with high-risk signals are not published. These are signals, not a guarantee that installing it is safe. Try it in a sandbox project first.
Sources
- AI Templates (aitmpl) first seen 13 Nov 2026 · last seen 18 Nov 2026 · its page on the source · figures reported by the source: aitmpl downloads 9
opus-5-5-advisor
Use Claude Opus 5.5 as the advisor your main model consults at decision points (before committing to an approach, on a recurring error, before declaring a task done). It can advise any main model except Fable (Haiku, Sonnet and Opus up to 5.5). Pinned to the full model ID, so it stays on 5.5. Requires Claude Code v2.1.280+ on the Anthropic API or v2.1.300+ on Claude Platform on AWS; on the Anthropic API, the advisor stays off when DISABLE_TELEMETRY or CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC is set.
As shown on www.aitmpl.com when we last read it