bo-faq-bot_skill

This skill builds and queries a FAQ knowledge base from markdown files, enabling automatic answers and fast, fuzzy matching.
  • Python

2.5k

GitHub Stars

3

Bundled Files

2 months ago

Catalog Refreshed

3 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 bo-faq-bot

  • _meta.json286 B
  • README.md1.6 KB
  • SKILL.md2.7 KB

Overview

This skill builds a searchable FAQ knowledge base from markdown files and provides tools to add, import, search, export, and manage FAQ entries. It matches incoming user questions to the best FAQ answer using fuzzy text matching and returns a confidence score for each match. It is designed for Feishu (Fieshu) business bots and simple CLI workflows.

How this skill works

The skill parses markdown files where H2 headings become questions and the following body becomes answers, then stores entries in a local knowledge base. It uses a TF-IDF-like fuzzy matching algorithm to score incoming queries and returns the top matches with a confidence score (0.0–1.0). CLI commands let you init, add, import, search, list, remove, export, and view stats; thresholds and top-N results are configurable.

When to use it

  • Set up an automatic FAQ responder in Feishu/Fieshu chatbots.
  • Create a lightweight knowledge base from existing markdown documentation.
  • Add or update FAQ entries without a database or heavy infrastructure.
  • Search and surface best-matching answers for incoming user questions.
  • Export or back up FAQs as markdown or JSON for portability.

Best practices

  • Format source files with clear H2 headings for each question and concise answer bodies for best parsing.
  • Tune the search threshold to balance precision and recall for your organization’s language.
  • Tag entries or keep a consistent question phrasing style to improve matching quality.
  • Regularly export or back up the knowledge base to avoid accidental data loss.
  • Test common user queries in the target chat environment and refine answers for clarity.

Example use cases

  • Import product or support FAQs from markdown documentation and serve answers automatically in Feishu.
  • Create a small business bot answering hours, payment methods, and account questions without a web backend.
  • Run periodic imports from a docs repo and export JSON for integration with other tools.
  • Use the search command in an automated agent flow: detect a user question, run search, and send the top answer if confidence is above threshold.
  • Maintain a curated FAQ list and remove outdated entries via the CLI remove command.

FAQ

Each H2 heading (## Question) becomes an FAQ question; the text following that heading until the next H2 is stored as the answer.

What does the confidence score mean?

Scores (0.0–1.0) reflect fuzzy matching strength; the default threshold is 0.3 and you can adjust it to control when automated replies are allowed.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational
bo-faq-bot skill by openclaw/skills | VeilStrat