tot-adaptable

Applies internal Tree-of-Thoughts branching with task-specific lenses to critique, discovery, planning, and review.

Updated Jun 29, 2026
One-click install
npx skills add https://github.com/thorsenk/skills --skill tot-adaptable-thorsenk
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: tot-adaptable
Source: https://github.com/thorsenk/skills/tree/main/skills/tot-adaptable
Command: npx skills add https://github.com/thorsenk/skills --skill tot-adaptable-thorsenk

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? First-impression answers often anchor on the most obvious frame and miss design intent, hidden tradeoffs, or alternative interpretations. This Skill reduces anchoring and confirmation bias by branching reasoning internally across multiple lenses before delivering a concise synthesis. ## Core Features & Use Cases - Internal Multi-Lens Branching: Classifies the task (critique, discovery, decision, plan, audit, review), selects 3-5 fitting lenses, branches privately, prunes weak branches, and merges results. - Charitable-First Critique Pattern: For specs, protocols, and minimal formats, analyzes decision, likely intent, enabled benefits, and tradeoffs before flagging real risks. - Concise Output Discipline: Hides the branching process, lists lenses only on request, and asks at most one high-value clarifying question when genuinely blocked. - Use Case: When reviewing a deliberately minimal API spec, use this Skill to infer why competent designers omitted features, identify genuine ambiguity risks, and recommend the smallest useful clarification instead of a long flaw list. ## Quick Start Use the tot-adaptable skill to analyze this plan with multiple internal lenses and recommend the smallest useful next step.

Frequently Asked Questions about tot-adaptable

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I use Tree of Thoughts prompting without exposing chain-of-thought?▼

Branch reasoning internally across 3-5 task-appropriate lenses, prune weak or duplicative branches, and present only the merged synthesis. The method stays private; the user sees a concise answer with the smallest useful next step.

What is multi-lens reasoning used for?▼

Multi-lens reasoning reduces anchoring by examining a problem from angles like design intent, disagreement, adoption cost, and user fit. It applies to critiques, discovery interviews, decisions, plans, audits, and reviews.

How do I critique a minimal spec without treating omissions as flaws?▼

Use a charitable-first pattern: identify the decision, infer likely intent, note what it enables, weigh the tradeoff, then propose the smallest useful clarification. Call out genuine risks plainly after that analysis.

When should I not use Tree-of-Thoughts branching?▼

Avoid it for simple, unambiguous tasks where one obvious frame suffices, since branching adds latency without improving the answer. It is designed for consequential or ambiguous work where anchoring would distort the result.

Does this approach work for discovery and user interviews?▼

Yes. For discovery tasks it swaps in empathy, vocabulary, boundary, assumption, and next-question lenses, and asks at most one high-value question when the answer is genuinely blocked.