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.

Jump to install

Repository: BochengYin/ai-wiki-toolkit
Last source activity: April 21, 2026 at 10:48
Detected SKILL.md language: English
Stars: 6
Forks: 0

Install options

The review-first prompt is selected by default. You can switch to a direct command or download a local copy.

Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.

A direct command skips the review prompt. Inspect the source before running it.

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

The command stays on one line. Scroll horizontally to inspect it before copying.

Prefer a local copy? Download the files currently available to SkillsMP.

Review the source files

Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.

File Explorer

6 files

SKILL.md

Source instructions · Read-only preview

Back to search Back to installation

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.