How do I install the Obsidian Search skill?

Run npx skillmds@latest add wilbeibi/obsidian-search in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Obsidian Search safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Obsidian Search?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Obsidian Search free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Obsidian Search?

wilbeibi (@wilbeibi) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Obsidian Search by wilbeibi — Agent Skill

Name: Obsidian Search
Author: wilbeibi

File contents

obsidian-search

Find Markdown notes and tasks in an Obsidian vault without loading the whole vault into context.

Routing

Topic / fuzzy / "notes about X" → scripts/meili_search.py. Typo-tolerant full-text over a Meilisearch index; best when the user won't remember exact words. Expand synonyms yourself across queries.
Exact content, frontmatter, tags, task glyphs → direct rg.
Natural-language dates, overdue comparisons, CJK term extraction → scripts/vault_search.py.

If Meilisearch is unreachable or unconfigured (script exits non-zero), fall back to tiers 2–3 — they only need the local vault.

Env: OBSIDIAN_VAULT (vault root), MEILI_URL (default http://127.0.0.1:7700), MEILI_SEARCH_KEY (file fallback ~/meilisearch/search-key.txt), MEILI_INDEX (default notes).

Meilisearch

Run with --help for all flags. Never dump raw hits into context — use --fields and small --limit.

uv run scripts/meili_search.py "spaced repetition" --limit 5 --fields "title,path"
uv run scripts/meili_search.py "prompt caching" --filter "tags=ai" --sort "updated:desc" --json

Hit path values are vault-relative (e.g. Library/Book Notes/Influence.md); join with $OBSIDIAN_VAULT to open.

Direct Patterns

rg "keyword" --glob "*.md" "$OBSIDIAN_VAULT"
rg -lU 'tags:\n(\s*- .*\n)*?\s*- tagname\b' --glob "*.md" "$OBSIDIAN_VAULT"  # YAML list tags; prefer meili --filter "tags=tagname" when index is up
rg "created: $(date +%Y-%m)" --glob "*.md" "$OBSIDIAN_VAULT"
rg "^- \[ \].*📅 $(date +%F)" --glob "*.md" "$OBSIDIAN_VAULT"
rg "^- \[ \].*⏫" --glob "*.md" "$OBSIDIAN_VAULT"

Date-Aware Script

uv run scripts/vault_search.py "recent week psychology"
uv run scripts/vault_search.py "overdue tasks"
uv run scripts/vault_search.py "找3-2-1技巧" --raw   # --raw: plain paths for piping

Notes

created: YYYY-MM-DD frontmatter drives dates; vault_search.py falls back to file mtime.
Task markers (Tasks plugin): 📅 due, 🛫 start, ⏳ scheduled, ⏫/🔼/🔽 priority.
Synonym expansion is the agent's job, not the scripts'.

---name: obsidian-searchdescription: Search an Obsidian vault for topics, exact text, tags, recent notes, and tasks. Use when fuzzy, date-aware, due, or overdue note retrieval is needed.---# obsidian-searchFind Markdown notes and tasks in an Obsidian vault without loading the whole vault into context.## Routing1. **Topic / fuzzy / "notes about X"** → `scripts/meili_search.py`. Typo-tolerant full-text over a Meilisearch index; best when the user won't remember exact words. Expand synonyms yourself across queries.2. **Exact content, frontmatter, tags, task glyphs** → direct `rg`.3. **Natural-language dates, overdue comparisons, CJK term extraction** → `scripts/vault_search.py`.If Meilisearch is unreachable or unconfigured (script exits non-zero), fall back to tiers 2–3 — they only need the local vault.Env: `OBSIDIAN_VAULT` (vault root), `MEILI_URL` (default `http://127.0.0.1:7700`), `MEILI_SEARCH_KEY` (file fallback `~/meilisearch/search-key.txt`), `MEILI_INDEX` (default `notes`).## MeilisearchRun with `--help` for all flags. **Never dump raw hits into context** — use `--fields` and small `--limit`.```bashuv run scripts/meili_search.py "spaced repetition" --limit 5 --fields "title,path"uv run scripts/meili_search.py "prompt caching" --filter "tags=ai" --sort "updated:desc" --json```Hit `path` values are vault-relative (e.g. `Library/Book Notes/Influence.md`); join with `$OBSIDIAN_VAULT` to open.## Direct Patterns```bashrg "keyword" --glob "*.md" "$OBSIDIAN_VAULT"rg -lU 'tags:\n(\s*- .*\n)*?\s*- tagname\b' --glob "*.md" "$OBSIDIAN_VAULT" # YAML list tags; prefer meili --filter "tags=tagname" when index is uprg "created: $(date +%Y-%m)" --glob "*.md" "$OBSIDIAN_VAULT"rg "^- \[ \].*📅 $(date +%F)" --glob "*.md" "$OBSIDIAN_VAULT"rg "^- \[ \].*⏫" --glob "*.md" "$OBSIDIAN_VAULT"```## Date-Aware Script```bashuv run scripts/vault_search.py "recent week psychology"uv run scripts/vault_search.py "overdue tasks"uv run scripts/vault_search.py "找3-2-1技巧" --raw # --raw: plain paths for piping```## Notes- `created: YYYY-MM-DD` frontmatter drives dates; `vault_search.py` falls back to file mtime.- Task markers (Tasks plugin): `📅` due, `🛫` start, `⏳` scheduled, `⏫`/`🔼`/`🔽` priority.- Synonym expansion is the agent's job, not the scripts'.

Frequently asked questions

Obsidian Search