Featured

Deploy OpenClaw in 60 seconds — 20% off logoDeploy OpenClaw in 60 seconds — 20% off

Launch OpenClaw on Hostinger in about 60 seconds and keep your agent live 24/7. Our referral link gives you 20% off, no coupon code needed.

Launch on Hostinger
Run your Hermes agent on Hostinger, fully managed logoRun your Hermes agent on Hostinger, fully managed

Launch Hermes on Hostinger in one click, fully managed, no VPS knowledge needed. Use code ZACAARON10 for 10% off.

Launch on Hostinger
Crawl and scrape any site into clean data, 10% off logoCrawl and scrape any site into clean data, 10% off

Firecrawl crawls and scrapes any site into clean markdown for your agent. Get 1,000 free credits, and new users get 10% off their first purchase.

Try Firecrawl free
6,000+ web scrapers for your AI agent, start free logo6,000+ web scrapers for your AI agent, start free

Apify gives your agent live web data: 6,000+ prebuilt scrapers and actors, MCP-ready. Sign up free with $5 in usage credits.

Try Apify free
One API to scrape, enrich, and extract the internet. logoOne API to scrape, enrich, and extract the internet.

Context.dev gives your agents a single API to scrape, enrich, and extract live web data — no proxies, no parsers, no maintenance.

Start building free
SetupClaw: done-for-you OpenClaw for founders & exec teams logoSetupClaw: done-for-you OpenClaw for founders & exec teams

White-glove OpenClaw for founders and exec teams (4–50+ employees): we install, harden, integrate your tools, and maintain it — secured from day one.

Get it set up for you
SEO data APIs for your agent, $1 free credit logoSEO data APIs for your agent, $1 free credit

DataForSEO gives your agent live access to SERP results, keyword data, backlinks, and on-page SEO data through one API. New accounts get a $1 credit, good for up to 20,000 keyword or backlink lookups.

Try DataForSEO free
Reach 48,000+ AI builders

A flat monthly placement in front of developers actively installing AI tools. No lock-in, cancel anytime.

Advertise here

Summary

Obsidian-based knowledge base via MCP. Auto-classified notes, cache-first lookup, cross-AP knowledge sharing.

Install to Claude Code

/plugin install kb-skill@kb-skill

Run in Claude Code. Add the marketplace first with /plugin marketplace add huangsron/kb-skill if you haven't already.

README.md

kb-skill

A Claude Code plugin that drives the obsidian-kb MCP server — a personal knowledge base backed by an Obsidian vault, shared across multiple apps/projects via 9 MCP tools.

What it does

When you ask a question or want to record something, this skill walks Claude through the right MCP tools in the right order:

1. kb_lookup — try the cache trigger index first; if hit, quote the answer verbatim. 2. kb_search — full-text search, scoped to the current project via tags derived from cwd. 3. kb_read — read full notes when search hits. 4. kb_create — auto-classify and persist result-shaped answers (bug fixes, configs, commands, API endpoints). Discussion/exploration notes ask first. 5. kb_log_append — keep a chronological trail.

Plus kb_update_frontmatter, kb_list_by_tag, kb_reindex_cache, kb_index_rebuild.

Prerequisites

You need the obsidian-kb MCP server running and reachable. See huangsron/obsidianMcp. Quick version:

git clone https://github.com/huangsron/obsidianMcp
cd obsidianMcp
cp .env.example .env  # fill in MCP_AUTH_TOKEN (openssl rand -hex 32)
docker compose up -d

The plugin assumes the server is at http://127.0.0.1:8787/mcp.

Install

# 1. Add this repo as a Claude Code plugin marketplace
/plugin marketplace add huangsron/kb-skill

# 2. Install the plugin
/plugin install kb-skill@kb-skill

After install, you must create .mcp.json with your token, because tokens are not stored in the repo. Find the installed plugin directory (usually ~/.claude/plugins/cache/<marketplace>/<plugin>/), then:

cp .mcp.json.example .mcp.json
# Edit .mcp.json — replace <YOUR_TOKEN_HERE> with the value of MCP_AUTH_TOKEN from your obsidian-kb server's .env

Restart Claude Code. Verify with:

/plugin

You should see kb-skill listed and the obsidian-kb MCP server connected.

Usage

Trigger the skill in two ways:

  • Slash command: /kb followed by what you want to do
  • Direct phrases: "查知識庫…", "記到知識庫…", "kb lookup…", "kb search…"

The skill stays out of the way on generic phrases like "查一下" — those are intentionally NOT triggers.

Cross-AP context

The KB is shared across projects. The skill auto-derives an AP tag from basename(cwd) and uses it as a filter in kb_search:

  • cwd = D:\work\local\sron\obsidianMcp → search with tags: ["obsidianMcp"]
  • cwd = D:\work\local\sron\tes → search with tags: ["tes"]

If a search returns nothing, it widens automatically (drops the tag filter).

What gets auto-saved vs asked

| Kind | Behavior | |------|----------| | Bug fixes, configs, commands, API endpoints | Auto kb_create | | Discussions, trade-offs, exploration | Ask first | | Chit-chat / topic already in vault | Don't save |

Updating

/plugin marketplace update kb-skill
/plugin update kb-skill@kb-skill

Files

kb-skill/
├── .claude-plugin/
│   ├── plugin.json           # plugin manifest
│   └── marketplace.json      # marketplace index (single-plugin)
├── .mcp.json.example         # copy to .mcp.json after install, fill token
├── skills/kb/SKILL.md        # the actual skill instructions
└── README.md

.mcp.json is gitignored; tokens never enter version control.

License

MIT

Related plugins

Browse all →