Communitygithub.com

TIKAZI/TIKAZ-Codex-Knowledge-Research

Traceable research and decisions with evidence ledgers, disagreement, confidence, and knowledge feedback.

TIKAZ-Codex-Knowledge-Research 是什麼?

TIKAZ-Codex-Knowledge-Research is a Codex agent skill that traceable research and decisions with evidence ledgers, disagreement, confidence, and knowledge feedback.

相容平台~Claude CodeCodex CLI~Cursor
npx skills add TIKAZI/TIKAZ-Codex-Knowledge-Research

在你喜歡的 AI 中提問

開啟一個已預先載入此 Agent Skill 的新對話。

說明文件

Knowledge and Research

Designed, integrated, refactored, and continuously maintained by TIKAZ.

TIKAZ method

  1. Define the question, decision, time boundary, source types, and evidence threshold.
  2. Convert source files with document-to-markdown only when needed; conversion is not verification.
  3. Choose one owner: academic-research, product-strategy, or solo-business-operations.
  4. Retrieve personal knowledge with personal-knowledge-memory before important reusable work, but treat it as candidate historical context.
  5. Build an evidence ledger with source, date, claim, support, contradiction, confidence, and relevance.
  6. Record knowledge adoption only when it changes execution; record success only with user or objective evidence.

Completion gate

Require a reproducible source trail, visible uncertainty, contrary evidence, no invented citations, a decision or synthesis matched to the original question, and a list of claims that need future revalidation.

Read references/routing.md and references/output-contract.md.

相關技能