agent-testing-bot

Automates end-to-end bot testing across Discord, Slack, Telegram, WeChat, Lark, QQ, and iMessage on macOS.

74|11|Updated Jul 4, 2024
One-click install
npx skills add https://github.com/OpenSourceAGI/qwksearch-research-agent --skill agent-testing-bot-opensourceagi
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-testing-bot
Source: https://github.com/OpenSourceAGI/qwksearch-research-agent/tree/main/apps/qwk-in-lobe/.agents/skills/agent-testing-bot
Command: npx skills add https://github.com/OpenSourceAGI/qwksearch-research-agent --skill agent-testing-bot-opensourceagi

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve? Verifying chat bot behavior across real messaging apps normally requires manual clicking, typing, and screenshotting in each native client. This Skill automates that end-to-end verification by driving the actual desktop apps (Discord, Slack, Telegram, WeChat, Lark, QQ) via osascript, or the iMessage BlueBubbles bridge, and capturing screenshot evidence of bot responses. ## Core Features & Use Cases - Cross-platform bot drivers: Per-platform shell scripts activate the native app, navigate to a channel or contact, send a message (clipboard-based for CJK support), wait, and capture the response window. - iMessage bridge regression testing: A dedicated script exercises the Electron imessageBridge.* IPC handlers, the local bridge HTTP server, and the BlueBubbles REST API, asserting config save, webhook registration, and secret enforcement. - Acceptance pipeline integration: Extends the generic acceptance skill's PLAN → EXECUTE → FINISH flow with report generation and ingestion. - Use Case: After changing how your LobeHub bot replies in WeChat, run the WeChat driver to send a test message to a contact and automatically screenshot the bot's reply for visual verification. ## Quick Start Ask the agent to test the bot in Discord by sending a message to the bot-testing channel and capturing the response screenshot.

Frequently Asked Questions about agent-testing-bot

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

FAQPage Schema
How do I test a Discord bot automatically on macOS?

Run the test-discord-bot.sh script with a channel name and message. It activates Discord via osascript, navigates with the Cmd+K Quick Switcher, pastes the message from the clipboard, waits for the reply, and captures a screenshot of the response.

How to send messages to WeChat or QQ bots from a script?

Use the test-wechat-bot.sh or test-qq-bot.sh scripts with a contact name and message. They search via Cmd+F, paste text through the clipboard to support CJK characters, send with Enter, and screenshot the result after a configurable wait.

Can this bot testing run headless or in cloud CI?

No. Bot-channel testing is macOS-only because it drives real native apps with osascript and captures evidence with OS-level screencapture. It requires a logged-in desktop session, accessibility permission, and an awake, unlocked display.

Why are my bot test screenshots completely black?

Black screenshots mean macOS Screen Recording (TCC) permission is missing or the display is asleep, locked, or on a screensaver. Run the check-screen-recording.sh gate first and keep the display awake with caffeinate during capture.

How is iMessage bot testing different from the other platforms?

iMessage has no osascript-driven app; it uses a BlueBubbles bridge. The test-imessage-bridge.sh script exercises Electron IPC handlers, the local bridge HTTP server, and the BlueBubbles REST API, asserting config save, webhook registration, and secret enforcement.