WhatsApp FAQ Bot

Build and query a FAQ knowledge base from markdown files. Use when asked to create a FAQ bot, set up automatic answers, build a knowledge base, add FAQ entri...

Security Scan

The skill's code, instructions, and requirements are internally consistent with a local FAQ/markdown-based knowledge base and do not request unrelated credentials or network access.

Purpose & Capability

The name/description (WhatsApp FAQ Bot) match the included code and SKILL.md: the script builds a local FAQ KB from markdown, supports search, import, export, and storage. There are no unexpected requirements (no cloud APIs, no unrelated binaries). Storing data under ~/.faq-bot (or overridden by FAQ_BOT_DIR) is appropriate for this purpose.

Instruction Scope

SKILL.md only instructs the agent to run the included CLI (init, add, import, search, list, export, remove, stats). The instructions reference reading user-supplied markdown files and operating on a local knowledge base; they do not direct the agent to read unrelated system files, exfiltrate data, or contact external endpoints. The SKILL.md suggests integration (cron/agent prompt) but gives no steps that broaden scope beyond the local CLI.

Install Mechanism

There is no install spec (instruction-only skill) and the package includes a single plain-Python script. No remote downloads, package managers, or archive extraction are required. This minimizes installation risk.

Credentials

The registry declares no required environment variables, and the skill requires no credentials — appropriate. The code and README do reference an optional FAQ_BOT_DIR environment variable to override the storage path; this is reasonable but is an undeclared optional env var (not a secret). It's a minor inconsistency between declared requires.env (none) and optional env usage in code/README.

Persistence & Privilege

The skill does not request always:true, does not modify other skills or global agent settings, and persists only its own knowledge-base JSON under the user's home directory (or an overridden directory). This scope of persistence is proportional to a local FAQ tool.

Assessment

This skill appears to do exactly what it says: build and search a local FAQ knowledge base from markdown. It does not contact external services or request credentials. Notes before installing: (1) Data is stored on disk by default at ~/.faq-bot (changeable with FAQ_BOT_DIR) — avoid storing sensitive secrets in FAQs or choose a protected directory. (2) The README/script mention an optional FAQ_BOT_DIR env var but the skill metadata lists no required env vars — this is harmless but a small mismatch to be aware of. (3) When integrating into an automated agent or cron job, ensure the agent runs in an environment with appropriate file permissions so your FAQ data isn't exposed. If you need the skill to access remote WhatsApp APIs or external services, those integrations are not present and would require additional secure credentials and review.

Build a searchable knowledge base from markdown files. Match incoming questions to the best FAQ answer using fuzzy text matching.

Quick Start

# Initialize a new FAQ knowledge base
python scripts/faqbot.py init

# Add a FAQ entry interactively
python scripts/faqbot.py add -q "What are your business hours?" -a "We are open Monday to Friday, 9 AM to 6 PM CET."

# Import FAQs from a markdown file
python scripts/faqbot.py import faq-source.md

# Search for the best matching answer
python scripts/faqbot.py search "when are you open"

# List all FAQ entries
python scripts/faqbot.py list

# Export all FAQs to markdown
python scripts/faqbot.py export --format md -o faqs-export.md

# Export as JSON
python scripts/faqbot.py export --format json -o faqs.json

# Remove a FAQ entry by ID
python scripts/faqbot.py remove 3

# Get stats about the knowledge base
python scripts/faqbot.py stats

Commands

Command	Args	Description
init		Create a new empty knowledge base
add	-q QUESTION -a ANSWER [-t TAGS]	Add a single FAQ entry
import	<file.md>	Import FAQs from markdown (H2 = question, body = answer)
search	<query> [--top N] [--threshold T]	Find best matching answer(s)
list	[--tag TAG]	List all FAQ entries
remove	<id>	Remove a FAQ entry
export	[--format md|json] [-o FILE]	Export knowledge base
stats		Show knowledge base statistics

Markdown Import Format

## What are your business hours?
We are open Monday to Friday, 9 AM to 6 PM CET.
Weekend support is available via email only.

## How do I reset my password?
Go to Settings > Account > Reset Password.
You will receive an email with a reset link.

## What payment methods do you accept?
We accept:
- Credit/debit cards (Visa, Mastercard)
- PayPal
- Bank transfer (EU only)

Each H2 heading becomes a question, the body below becomes the answer.

Search Scoring

• Uses TF-IDF-like fuzzy matching on question text
• Returns confidence score (0.0 to 1.0)
• Default threshold: 0.3 (adjustable with --threshold)
• Returns top 3 matches by default (adjustable with --top)

Integration with OpenClaw

This skill is designed to work as a WhatsApp FAQ bot. When a user asks a question, the agent can use the search command to find the best match and respond automatically. Configure it in your cron or agent system prompt.