obsidian

تایید شده

Work with Obsidian vaults as a knowledge base. Features: fuzzy/phonetic search across all notes, auto-folder detection for new notes, create/read/edit notes with frontmatter, manage tags and wikilinks. Use when: querying knowledge base, saving notes/documents, editing existing notes by user instructions.

@openclaw
MIT۱۴۰۴/۱۲/۳
رد شده
(0)
۱.۰k
۳۱۷
۳۶۳

نصب مهارت

مهارت‌ها کدهای شخص ثالث از مخازن عمومی GitHub هستند. SkillHub الگوهای مخرب شناخته‌شده را اسکن می‌کند اما نمی‌تواند امنیت را تضمین کند. قبل از نصب، کد منبع را بررسی کنید.

نصب سراسری (سطح کاربر):

npx skillhub install openclaw/skills/obsidian

نصب در پروژه فعلی:

npx skillhub install openclaw/skills/obsidian --project

مسیر پیشنهادی: ~/.claude/skills/obsidian/

بررسی هوش مصنوعی

رد شده
این مهارت معیارهای کیفیت را ندارد

محتوای SKILL.md

---
name: obsidian
description: >
  Work with Obsidian vaults as a knowledge base. Features: fuzzy/phonetic search across all notes,
  auto-folder detection for new notes, create/read/edit notes with frontmatter, manage tags and wikilinks.
  Use when: querying knowledge base, saving notes/documents, editing existing notes by user instructions.
---

# Obsidian Knowledge Base

Obsidian vault = folder with Markdown files + `.obsidian/` config.

## Configuration

- **Vault Path:** `/home/ruslan/webdav/data/ruslain`
- **Env:** `OBSIDIAN_VAULT=/home/ruslan/webdav/data/ruslain`

## CLI Usage

Scripts location: `/home/ruslan/.openclaw/workspace/skills/obsidian/scripts`

Note: Global flags (`--vault`, `--json`) must come BEFORE the command.

```bash
export OBSIDIAN_VAULT=/home/ruslan/webdav/data/ruslain
cd /home/ruslan/.openclaw/workspace/skills/obsidian/scripts

# Search (fuzzy/phonetic) - uses ripgrep for speed
python3 obsidian_search.py "$OBSIDIAN_VAULT" "query" --limit 10 --json

# List notes
python3 obsidian_cli.py --json list                    # all notes
python3 obsidian_cli.py --json list "Projects"         # in folder

# List folders
python3 obsidian_cli.py --json folders

# Read note
python3 obsidian_cli.py --json read "Note Name"

# Create note
python3 obsidian_cli.py --json create "Title" -c "Content" -f "Folder" -t tag1 tag2
python3 obsidian_cli.py --json create "Title" -c "Content" --auto-folder  # auto-detect folder

# Edit note
python3 obsidian_cli.py --json edit "Note" append -c "Text to append"
python3 obsidian_cli.py --json edit "Note" prepend -c "Text at start"
python3 obsidian_cli.py --json edit "Note" replace -c "New full content"
python3 obsidian_cli.py --json edit "Note" replace-section -s "Summary" -c "New section text"

# Tags
python3 obsidian_cli.py --json tags

# Links (incoming/outgoing)
python3 obsidian_cli.py --json links "Note Name"

# Suggest folder for content
python3 obsidian_cli.py --json suggest-folder "content text" --title "Note Title"
```

## Workflow: Query Knowledge Base

1. Run `obsidian_search.py` with user query
2. Read top results if needed for context
3. Formulate answer based on found content
4. Cite sources: `[[Note Name]]`

## Workflow: Save Note

1. If no folder specified → run `suggest-folder` or use `--auto-folder`
2. Create note with `create` command
3. Add appropriate tags based on content
4. Report created path to user

## Workflow: Edit Note by Prompt

User prompts like:
- "Добавь резюме в конец заметки X" → `edit X append -c "..."`
- "Перепиши заметку Y более кратко" → read note, rewrite, `edit Y replace -c "..."`
- "Добавь секцию 'Выводы' в заметку Z" → `edit Z replace-section -s "Выводы" -c "..."`

## Note Format

```markdown
---
created: 2024-01-15T10:30:00
modified: 2024-01-15T12:00:00
tags:
  - project
  - work
---

# Title

Content with [[wikilinks]] and #inline-tags.
```

## Wikilinks

- `[[Note Name]]` — link to note
- `[[Note Name|Display Text]]` — link with alias
- `[[Note Name#Section]]` — link to section

## Frontmatter Fields

Standard fields:
- `created` — creation timestamp
- `modified` — last edit timestamp  
- `tags` — list of tags
- `aliases` — alternative names for linking

## Search Behavior

`obsidian_search.py` uses:
- ripgrep for fast initial filtering
- Title matching (highest weight)
- Tag matching
- Fuzzy content search with phonetic transliteration (RU↔EN)
- Returns: path, title, score, matched context