witt3rd/zk-memory/zettelkasten-maintenance

Claude Code and agent skill by witt3rd/zk-memory

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Maintain a zk-memory corpus: tend (linlink), tend-writes (gardener), split, sidecar paths, FAISS embedder pin. Use when links may be broken, a note grew into a biography, the vector pin drifted, or zk-memory tend / tend-writes / split.

Install

npx vskill@latest install witt3rd/zk-memory/zettelkasten-maintenance

SKILL.md

zettelkasten-maintenance

Repo AGENTS.md is the contract. This file is the maintenance surface.

Link integrity — tend

zk-memory --root <root> tend check
zk-memory --root <root> tend mint --write
zk-memory --root <root> tend repair --write
zk-memory --root <root> tend robustify --write

Dry-run without --write. Actions: check / repair / mint / robustify. tend walks up from the corpus root to find linlink.toml. linlink is the tool; darnlink is the old name.

Gardener — tend-writes (needs an LLM)

Reconciles recent writes (by mtime): fold a duplicate into an existing note (append-only) and retire the original to <root>/.archive/ (reversible, never deleted), or append out-links. decision notes never merge. Capture fast, integrate later.

De-merge — split (needs an LLM)

split_note is the inverse of merge: summary parent + atomic children, original retired to .archive/. Not the authoring "branch a new sibling and keep the old note." split-candidates is a mechanical size sweep (no LLM). tend-writes --split-sweep N may split only notes that sweep listed.

Sidecar

Everything planted besides notes lives under <root>/.zk/ (lock, FTS index, vector pin, ingest, trace). Never beside the corpus in the parent. Retired notes: <root>/.archive/.

Vector pin

<root>/.zk/index/vector/meta.json records model + dims. A later search with a different embedder raises EmbedderMismatch unless --rebuild-index. The index is derived and rebuildable. Committing it is a corpus policy (embeddings cost money), not a library rule.

CLI embedder is injected, not a backend string:

zk-memory --root <root> --embed-model <id> --embed-base <url> --embed-key <key> search "…"

Env: ZK_MEMORY_EMBED_MODEL / ZK_MEMORY_EMBED_BASE / ZK_MEMORY_EMBED_KEY. Any OpenAI-compatible /embeddings endpoint.

Ingest vs maintenance

ingest skips any source path whose components start with . (so .meta/ never inventories). That is an ingest limitation, not a tend rule — write those notes with write / merge. See the ingest skill.

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
0

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/witt3rd/zk-memory/zettelkasten-maintenance/badge)](https://verified-skill.com/skills/witt3rd/zk-memory/zettelkasten-maintenance)

Something wrong with this skill or its page? Tell us.