moltflow-ai_skill

This skill helps you automate WhatsApp with AI-powered replies, voice transcription, and knowledge base integration for faster, consistent customer

Python

2.6k

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill openclaw/skills --skill moltflow-ai

SKILL.md13.0 KB

Overview

This skill adds AI-powered WhatsApp automation: voice transcription, RAG knowledge base ingestion/search, style-profile training, and context-aware reply generation. It helps teams scale customer messaging with searchable business data and customizable tone. Use it to transcribe audio, upload documents, train writing styles, and generate safe, RAG-backed replies.

How this skill works

The skill exposes AI endpoints under /ai to queue voice transcriptions, index documents into a semantic knowledge base, train style profiles from chat history, and generate reply suggestions. Transcription and training run asynchronously (Celery tasks). Reply generation combines conversation context, retrieved RAG sources, and an applicable style profile to produce grounded, stylistically consistent responses.

When to use it

Transcribe incoming WhatsApp voice messages into text
Upload PDFs or TXT files to create a searchable knowledge base
Perform semantic search across uploaded documents for quick answers
Train a style profile to make AI replies match your team's tone
Generate auto-replies or preview suggested responses using RAG and style

Best practices

Enable tenant conversation context in Settings > Account > Data Access before using context-aware features
Use X-API-Key or Bearer token in every request and scope the key to ai read/manage
Upload clean, well-structured PDFs or text files (<=100MB) for best RAG results
Train per-chat profiles for personalized replies and a general profile for fallback
Preview responses before sending when using approval-required or high-risk intents

Example use cases

Queue voice message transcription and fetch the transcript to log customer requests
Ingest a product catalog PDF and run semantic queries for support answers
Train a Sales style profile from a sales chat and generate pitch-consistent replies
Preview an AI reply that uses RAG sources to cite policy excerpts
Delete outdated knowledge sources or style profiles when documents change

FAQ

Provide MOLTFLOW_API_KEY (or Bearer token) on each request. Base URL is https://apiv2.waiflow.app/api/v2 and all AI endpoints live under /ai.

Which plan is required for AI features?

AI features require Pro or higher. Voice transcription, RAG, style profiles, and reply generation are available on Pro; Business removes some quotas.