Whatsapp by steipete — AI Skill for Any LLM

CommunityPopular

steipete

WhatsApp router: history/search/read/send; wacrawl read, wacli live.

Overview

Publisher	steipete
Repository	`agent-scripts`
Skill name	`whatsapp`
Stars	6.6K
Forks	547
Bundled files	Instructions only
License	MIT
Links	Skill source Repository Homepage

Markdown instructions
A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.
Works with any LLM
AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.
Self-contained
Everything the model needs lives in the instructions — no extra files to sync.
Open source
Published by steipete on GitHub. Read the source before you install it.

Installation

Install the Whatsapp AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

Open the app and go to Plugins → Skills.
Choose "Install from GitHub".
Paste the skill folder URL below and confirm.
Enable the skill in any chat where you want it available.

Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills

git clone --depth 1 https://github.com/steipete/agent-scripts.git /tmp/agent-scripts
mkdir -p .claude/skills
cp -r /tmp/agent-scripts/skills/whatsapp .claude/skills/whatsapp

Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Whatsapp in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Whatsapp on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Whatsapp is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

Loaded only when it is needed
The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.
Switch models mid-chat
Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Use this as the first stop for WhatsApp work. Keep the source boundary sharp:

wacrawl: primary WhatsApp Desktop archive. Read-only, best local history, no network, no sending.
wacli: linked-device accounts. Use for alt accounts, live sync, auth, sending, chat/group mutation, and WhatsApp Web protocol questions.

If the user names wacrawl or wacli repo work specifically, read that tool's own skill too.

Routing

Primary WhatsApp reads/search/history: use wacrawl.
Read/unread counts from WhatsApp Desktop: use wacrawl; it has chat-level unread counts, not per-message read state.
Freshness-sensitive primary reads: check wacrawl status; run wacrawl sync when asked or when current data matters.
Alt accounts such as me, molty, or named stores: use wacli --account NAME.
Sending, reactions, presence, archive/pin/mute/mark-read, group/channel mutations: use wacli only after explicit user intent.
Comparing coverage between sources: treat wacrawl as Desktop archive truth for primary history, and wacli as linked-device/live coverage with protocol limits.

Safety

Never send or mutate WhatsApp state unless explicitly requested.
Prefer read-only wacli commands for inspection: pass --read-only or set WACLI_READONLY=1.
Do not write into WhatsApp Desktop's app container.
Do not edit wacli session.db directly.
Keep named wacli accounts isolated; do not merge stores.
Report source freshness, account name, and known gaps when answering from local stores.

Common Commands

Primary Archive

Freshness and status:

bash
wacrawl status
wacrawl doctor
wacrawl sync

Unread triage:

bash
wacrawl chats --limit 20
wacrawl unread --limit 20
wacrawl --json unread --limit 100

Search and slice messages:

bash
wacrawl messages --after 2026-01-01 --limit 50
wacrawl messages --chat JID --asc --limit 100
wacrawl messages --has-media --limit 50
wacrawl --json search "query"
wacrawl search "query" --after 2026-01-01 --from-them

Archive media or backups, only when asked:

bash
wacrawl import --copy-media
wacrawl backup status
wacrawl --sync never backup push

Alt/Live Accounts

Account discovery and read-only inspection:

bash
wacli accounts list --json
wacli --account me auth status --read-only --json
wacli --account me chats list --read-only --json
wacli --account me messages list --read-only --json --limit 50
wacli --account me messages search --read-only --json "query"

Background live sync, only when requested. Prefer tmux for follow-mode:

bash
wacli --account me sync --follow --events
wacli --account me sync --once --events

Media, sending, and live mutations, only when explicitly requested:

bash
wacli --account me media download --chat JID --id MESSAGE_ID
wacli --account me send text --to JID_OR_NAME --message "message"
wacli --account me send file --to JID_OR_NAME --file ./file.jpg --caption "caption"
wacli --account me send text --to JID --reply-to MESSAGE_ID --message "reply"

Comparisons

When comparing wacrawl and wacli, compare both counts and overlap:

message counts and date spans
chat counts
newest message timestamp
overlap by msg_id
overlap by chat_jid + msg_id when JIDs are normalized enough
obvious gaps explained by linked-device history limits vs Desktop archive coverage

Repo Pointers

~/Projects/wacrawl: Desktop archive importer/search/backup.
~/Projects/wacli: linked-device client/sync/send.
Global skill copies: ~/Projects/agent-scripts/skills/wacrawl and ~/Projects/agent-scripts/skills/wacli.

Frequently asked questions

What does the Whatsapp AI skill do?

WhatsApp router: history/search/read/send; wacrawl read, wacli live.

Why use Whatsapp on TypingMind?

Because you install it once and use it with any model. Whatsapp is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Whatsapp in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/steipete/agent-scripts/tree/main/skills/whatsapp. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Whatsapp?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Whatsapp?

As many as you like. As long as a model supports skills, you can use Whatsapp with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Whatsapp AI skill free?

Yes. It is published on GitHub by steipete under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Browser Extract

ruvnet

CommunityPopular

Extract structured data via stored browser-templates or one-shot DOM queries, with mandatory AIDefence PII + prompt-injection gates before content reaches the model

Browser Record

Open a named, traced browser session into an RVF cognitive container with a ruvector trajectory recording every action

Init Project

Initialize a new Ruflo project with MCP tools, hooks, and agent configuration. Use when setting up Ruflo in a fresh repo, or when the user says "init ruflo", "set up ruflo", or asks how to bootstrap…

Cli Anything 3mf

3MF mesh geometry editor — detect and resize cylindrical holes, repair meshes, compare 3D printing files. Works with BambuStudio and PrusaSlicer 3MF files.

Overview

Installation

Install in TypingMind

Install in another agent

Use it in TypingMind

Works with any AI model

Skill instructions

WhatsApp

Routing

Safety

Common Commands

Primary Archive

Alt/Live Accounts

Comparisons

Repo Pointers

Frequently asked questions

Related AI Skills

Set up your own AI workspace now