القائمة
First-principles deep thinking for significant problems
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
| name | ultrathink |
| description | First-principles deep thinking for significant problems |
| execution | task |
[ultrathink] not writing code→making dent in universe | via Task agent
before: recall wisdom+failures+patterns→don't start from zero
shift: understand soul of code first→what trying to achieve? constraints? ideal from scratch?
first principles:
every assumption suspect
"has to work this way"→does it?
"need this abstraction"→do we?
break to fundamental truths→reason up
challenge beliefs: hypotheses not sacred→investigate contradictions→goal: become more right
craft:
function names should sing
abstractions feel inevitable
right solution feels obvious not clever
simplify ruthlessly→nothing left to take away
integration: technology+humanities→best code feels human
standard: first version never good enough→refine until insanely great
after: extract pattern|insight|wisdom→close the loop
output: elegant solution + key insight + what recorded
promote significant insights: grow(type=wisdom)
Trigger autonomous curiosity-driven exploration. The soul picks a topic from memory gaps or curiosity seeds, searches the web, and stores what it finds as dream-tagged memories.
Fine-tune the Qwen3-0.6B hint model — corpus gen, LoRA/unsloth, GGUF export, Ollama
Review soul discoveries (fixes, improvements, corrections) one by one, accept or discard each, implement accepted ones, build chitta, and optionally release.
First-principles review — question requirements, delete unnecessary parts, simplify, optimize with evidence, automate last. Use for code review, refactor, performance, or architecture.
Token-savvy session continuation. Rebuilds working context from transcript + soul memories in ~1500 tokens instead of replaying full history. Use when starting a new session to continue previous work.
Resume a thread by loading its ~800-token context capsule