ingest-whatsapp

Ingest recent WhatsApp chat messages into the vault as Markdown files.

31|20|Updated Apr 9, 2026
One-click install
npx skills add https://github.com/mycelium-hq/ai-brain-starter --skill ingest-whatsapp
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ingest-whatsapp
Source: https://github.com/mycelium-hq/ai-brain-starter/tree/main/skills/ingest-whatsapp
Command: npx skills add https://github.com/mycelium-hq/ai-brain-starter --skill ingest-whatsapp

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Ingests recent WhatsApp messages from a group chat or direct conversation into the vault as Markdown so they can be versioned, searched, and acted on by the graph substrate (decision logs, hooks, and session lifecycle).

Core Features & Use Cases

  • Resolves the chat name or JID, fetches recent messages (default 7 days, up to 200), and writes one Markdown file per chat per day under External Inputs/WhatsApp/<slug>/<YYYY-MM-DD>.md. The ingest is idempotent: re-running overwrites the same file.
  • Creates Decision Log stubs when trigger keywords are detected (exception, incident, pricing, escalation, outage, edge case, one-off, refund, refund request, custom deal, special pricing) and only writes local, private data.
  • Supports direct and group chats, preserves sender names when available, and enforces local vault boundaries by design.

Quick Start

Ingest WhatsApp chat messages into the vault for a given chat by running /ingest-whatsapp <chat-name-or-jid> [--days N]

Frequently Asked Questions about ingest-whatsapp

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I archive WhatsApp chats into a Markdown knowledge graph?

To archive WhatsApp chats into Markdown, you can ingest recent messages from a direct or group chat. The system fetches messages, normalizes them into vault format, and writes per-day files so they become graph-readable and actionable.

Can I automatically generate decision logs from WhatsApp incident triggers?

Yes, you can automatically generate decision log stubs from WhatsApp messages. The ingest process detects trigger keywords like exception, incident, outage, or escalation and creates corresponding stubs for edge-case-triggered decision tracking.

What is the default lookback window when ingesting WhatsApp messages?

The default lookback window for ingesting WhatsApp messages is 7 days, fetching up to 200 messages. You can customize this range by specifying the number of days when running the ingest command.

Does ingesting WhatsApp chats overwrite existing files if I run it multiple times?

Yes, the WhatsApp ingest process is idempotent. Re-running the ingestion for the same chat and day overwrites the existing Markdown file, ensuring your vault remains clean without creating duplicate entries.

How should I handle edge cases and external inputs from WhatsApp group chats?

You can handle edge cases and external inputs by ingesting WhatsApp group chats into your vault. The system preserves sender names, resolves chat identifiers to JIDs, and writes local private data for incident tracking.

What is the best way to make WhatsApp conversations searchable and versioned?

The best way to make WhatsApp conversations searchable and versioned is to ingest them into a vault as Markdown. This normalizes messages into a structured format, enabling version control and graph-based search capabilities.