ai-wiki-consolidate-drafts

Use when handle-local AI wiki drafts have grown into a noisy backlog and related notes should be clustered, refined, or promoted without directly rewriting shared docs by default.

Ir para a instalação

Repositório: BochengYin/ai-wiki-toolkit
Última atividade na origem: 21 de abril de 2026 às 10:48
Idioma detectado do SKILL.md: inglês
Estrelas: 6
Forks: 0

Opções de instalação

Por padrão, está selecionado o prompt que primeiro revisa a origem. Você pode mudar para um comando direto ou baixar uma cópia local.

Instalar com Codex ou Claude Copie este prompt, cole no Codex, Claude ou outro assistente e deixe que ele revise a página da skill e instale para você.

Um comando direto ignora o prompt de revisão. Verifique a origem antes de executá-lo.

npx skills add https://github.com/BochengYin/ai-wiki-toolkit --skill ai-wiki-consolidate-drafts

O comando permanece em uma só linha. Role horizontalmente para revisá-lo antes de copiar.

Prefere uma cópia local? Baixe os arquivos disponíveis atualmente no SkillsMP.

Revise os arquivos de origem

Leia o SKILL.md e os arquivos complementares exibidos pelo SkillsMP antes de decidir se vai instalar.

Explorador de arquivos

6 arquivos

SKILL.md

Instruções da origem · Visualização somente leitura

Voltar à pesquisa Voltar à instalação

AI Wiki Consolidate Drafts

Use this skill when ai-wiki/people/*/drafts/ has accumulated related notes and the next step is to consolidate them into clearer draft clusters, promotion candidates, or explicit conflicts and supersessions.

Workflow

Read the relevant draft set, usually under one handle in ai-wiki/people/*/drafts/.

Read nearby context that may confirm or narrow the consolidation target:

ai-wiki/conventions/index.md
ai-wiki/decisions.md
ai-wiki/review-patterns/index.md
ai-wiki/problems/index.md
ai-wiki/features/index.md
relevant files under ai-wiki/trails/ when chronology matters
ai-wiki/metrics/ and _toolkit/metrics/*.json only as weak signals

Cluster drafts by durable topic, repeated failure mode, or repeated rule, not by filename alone.

For each cluster, choose one next action using references/candidate-types.md.

Use references/promotion-targets.md to propose the best destination when a cluster is mature enough for shared or repo-level memory.

Use references/conflict-and-supersession.md to flag conflicts, refinements, or supersession instead of silently merging away differences.

Default to refining handle-local drafts first. Do not directly rewrite ai-wiki/conventions/*.md, ai-wiki/review-patterns/*.md, ai-wiki/problems/*.md, or ai-wiki/features/*.md unless the user explicitly asks for promotion or doc creation.

Emit the result using references/output-contract.md.

Rules

Cluster first, judge second.

Do not promote every repeated draft.

Metrics may support prioritization, but they are weak signals and must not be the sole basis for promotion.

Make the suggested target path explicit whenever the action is Refine draft or Promotion candidate.

If a cluster is still unstable, keep it local and explain what evidence is still missing.

Shared conventions and review patterns still require human confirmation before writing shared docs.