OpenClaw Shared Skills

A collection of skills and tools for OpenClaw agents.

Installation

1. Clone this repo
2. Add tools/ to your PATH:

   export PATH="/path/to/openclaw-shared/tools:$PATH"

3. Copy desired skills to your OpenClaw workspace:

   cp -r skills/resy ~/.openclaw/skills/

---

Skills

resy

Restaurant reservations via Resy API

Search restaurants, check availability, and book tables on Resy.

Capability	Description
resy search "name"	Find restaurants by name
resy browse YYYY-MM-DD	See what's available on a date
resy slots <venue_id> <date>	Get time slots for a restaurant
resy book	Complete a reservation
resy reservations	List upcoming reservations
resy cancel	Cancel a reservation

Setup required:

1. Log into resy.com
2. Open browser DevTools → Application → Cookies
3. Copy the authToken cookie value
4. Save to ~/.config/resy/auth_token

Tool: tools/resy (Python, no dependencies beyond stdlib)

---

opentable

Restaurant search via browser automation

Search and book restaurants on OpenTable. Uses OpenClaw browser — no API key needed.

Capability	Description
Search by name, cuisine, or neighborhood	Deep-links to search results
Check availability	See time slots for a specific restaurant
Book a table	Click through reservation flow

Setup required:

1. Log into opentable.com in your OpenClaw browser
2. That's it — uses your logged-in session

Tool: None (browser-based)

---

kindle

Send articles to Kindle

Fetches web articles, cleans them up, and emails to your Kindle as readable HTML.

Capability	Description
send-to-kindle <url>	Fetch article and send to Kindle

Setup required:

Create ~/.openclaw/.env:

SMTP_HOST=smtp.gmail.com
SMTP_PORT=587
SMTP_USER=you@gmail.com
SMTP_PASS=your-app-password
KINDLE_EMAIL=you_abc123@kindle.com

How to get these:

1. SMTP credentials: Use Gmail with an App Password (requires 2FA enabled)
2. KINDLE_EMAIL: Find in Amazon → Manage Your Content and Devices → Preferences → Personal Document Settings → Send-to-Kindle Email

Tool: tools/send-to-kindle (Python + Playwright)

---

imsg

iMessage/SMS on macOS

Read and send iMessages from your Mac.

Capability	Description
imsg chats	List recent conversations
imsg history --chat-id N	Read messages from a chat
imsg send --to "+1..." --text "hi"	Send a message
imsg watch --chat-id N	Stream incoming messages

Setup required:

1. Install: brew install steipete/tap/imsg
2. Grant Full Disk Access to your terminal (System Settings → Privacy)
3. Grant Automation permission for Messages.app when prompted

Tool: None (installed via brew)

---

qmd

Local markdown search

Search across markdown files using keyword or semantic search. Great for notes, docs, knowledge bases.

Capability	Description
qmd search "query" -c collection	Keyword search (fast)
qmd vsearch "query" -c collection	Semantic search (slower)
qmd get "path/to/file.md"	Retrieve a specific file

Setup required:

1. Install: bun install -g qmd
2. Index your markdown folder:

   qmd init my-notes ~/path/to/notes
   qmd update -c my-notes
   qmd embed -c my-notes  # enables semantic search

Tool: None (installed via bun)

---

text-processor

Auto-classify iMessages and take action

Watches incoming texts, classifies them using local LLM (DeepSeek), verifies with Claude, and takes action — saving memories, proposing reminders, or detecting calendar events.

Capability	Description
text-processor	Run the daemon (polls every 3s)
theo-send --chat-id N --text "msg"	Send as assistant + watch for replies

Architecture:

1. DeepSeek-R1:8b (local via Ollama) does fast first-pass classification
2. Claude (via OpenClaw) verifies before taking action
3. Memories saved to markdown files, reminders/calendar sent as confirmation texts

Dependencies:

Dependency	Install	Purpose
Python 3	(pre-installed on macOS)	Runtime
Ollama	brew install ollama	Local LLM server
DeepSeek-R1:8b	ollama pull deepseek-r1:8b	Classification model
imsg	brew install steipete/tap/imsg	iMessage access
OpenClaw	npm install -g openclaw	Claude verification + cron
requests	pip3 install requests	HTTP client for Ollama

Setup required:

1. Install dependencies above
2. Create config file ~/.config/text-processor/config.json:

{
  "my_number": "+1234567890",
  "bot_identifiers": ["your-bot@email.com"],
  "contacts_file": "/path/to/contacts.txt",
  "memory_dir": "/path/to/memory/friends",
  "prompt_file": "/path/to/classifier-prompt.txt",
  "watched_chats_file": "/path/to/watched-chats.json",
  "decisions_file": "~/.openclaw/text-processor-decisions.json",
  "state_file": "~/.openclaw/text-processor-state.json",
  "signature": "- YourBot (Your Assistant)"
}

3. Copy classifier-prompt.txt to your prompt_file path
4. Create a contacts file (format: +1234567890|Name per line)
5. Grant Full Disk Access to your terminal
6. Start Ollama: ollama serve
7. Run: text-processor

Tools: tools/text-processor, tools/theo-send, tools/classifier-prompt.txt (Python)

---

Quick Reference

Skill	What it does	Credentials needed
resy	Book restaurants on Resy	Resy auth token (from cookie)
opentable	Book restaurants on OpenTable	OpenTable login (in browser)
kindle	Send articles to Kindle	SMTP + Kindle email
imsg	Read/send iMessages	macOS permissions
qmd	Search markdown files	None
text-processor	Auto-classify texts, save memories	Config file + Ollama + OpenClaw