Use this skill to run end-to-end acceptance checks for bot channels in real desktop chat apps. It drives apps such as Discord, Slack, Telegram, WeChat, Lark, and QQ to send a message, wait for a response, and capture the result.
Use for real bot-channel acceptance in Discord, Slack, Telegram, WeChat/Weixin, Lark/Feishu, QQ or iMessage on macOS. Extends acceptance with native chat apps.
Run score Not tested
See how to installThis skill supports acceptance checks of bot channels in real chat apps. For Discord, Slack, Telegram, WeChat/微信, Lark/飞书, and QQ, it opens the app, navigates to a channel or contact, sends a message, waits for a reply, and captures the result window. iMessage is handled through a BlueBubbles bridge. The process has planning, execution, and finish phases, and its reporting and ingestion follow the generic acceptance skill.
Use it when a change affects a bot channel and you need to check the real conversation path in a native app, beyond what a web page or CLI can show. It supports checks that include native app interaction and a screen capture.
This skill is for macOS only. The target native app must be open and logged in. Screen capture requires macOS Screen Recording permission. It cannot run headlessly or in the cloud.
This overview was written by AI (OpenAI Codex) from the public SKILL.md. Check the original below for the exact behavior.
AI checked the SKILL.md for operations that need care and for setup requirements. The score comes only from actually running the skill (hands-on test).
Not counted in the score. Rated by AI (OpenAI Codex) as of Oct 26, 2026. Treat it as a guide and read the original before installing.
Review the instructions the agent will receive before you install.
This skill extends the generic acceptance skill (installed alongside at
../acceptance/) to the bot-channel surfaces. It does NOT replace the core
process: the same three phases apply —
PLAN (Steps 0–2) → EXECUTE (Steps 3–6) → FINISH (Step 7)
— and the same report + publish pipeline (result.json → report-init.sh →
lh acceptance run ingest … --source agent-testing). Read the acceptance skill's
SKILL.md for target grounding, the living logs, the Phase-1 approval gate, the
report format, and teardown. This file only adds the bot-channel surface.
For LobeHub environment/auth/probe specifics (dev server, seeded account, ports),
the adapter is .agents/acceptance/PROJECT.md.
| Change scope | Surface | Why |
|---|---|---|
| Bot channels (Discord / WeChat / Lark / …) | Native app via osascript / bridge | Only way to exercise the real channel end-to-end |
Route here only when the behavior under test is a bot channel. For CLI / Web / Electron changes, use the generic skill's surfaces.
Each platform folder has an index.md (activation, navigation, send-message,
verification snippets) and a test-<platform>-bot.sh driver.
| Platform | Guide | Quick switcher |
|---|---|---|
| Discord | discord/index.md | Cmd+K |
| Slack | slack/index.md | Cmd+K |
| Telegram | telegram/index.md | Cmd+F |
| WeChat / 微信 | wechat/index.md | Cmd+F |
| Lark / 飞书 | lark/index.md | Cmd+K |
| qq/index.md | Cmd+F | |
| iMessage | imessage/index.md | bridge (no osascript) |
Every osascript platform script shares one interface:
./$PLATFORM/test-$PLATFORM-bot.sh $CHANNEL_OR_CONTACT $MESSAGE [$WAIT_SECONDS] [$SCREENSHOT_PATH]
The script activates the app, navigates to the channel/contact, sends the
message, waits, and screenshots the result window (via the project adapter's
../../acceptance/scripts/capture-app-window.sh, which runs the installed
acceptance skill's screen-recording preflight). iMessage is the exception: it
uses a BlueBubbles bridge, not osascript — see imessage/index.md.
The osascript platforms drive real macOS apps (activate, keystroke, click, read accessibility, screenshot). Read the generic skill's general macOS-automation reference before a first bot run: ../../acceptance/references/osascript.md. The target native app must already be running and logged in.
Bot evidence is captured with OS-level screencapture (through
capture-app-window.sh), NOT CDP — so it comes out entirely black when macOS
Screen Recording (TCC) permission is missing OR the display is asleep / locked /
on a screensaver. Gate BEFORE any bot capture:
bash .agents/skills/acceptance/scripts/check-screen-recording.sh
Only exit 0 confirms both permission and a measured non-black frame. Missing tools or an undetermined check must block capture, not count as a pass.
Keep the display awake for the whole capture session (caffeinate -dimsu &, kill
when done). Because this surface depends on OS capture and native macOS apps,
bot channels are macOS-only — they do not run headless / in cloud.
No reviews yet. Tell others what you thought.
No login needed. Share how it worked for you, ask a question, or note where it did not run.
No comments yet. Be the first to write one.
Use this skill to verify a delivery by exercising the real product through a suitable surface such as a CLI, web app, desktop app, or iOS Simulator. It helps create or find a verification plan, capture evidence, and publish an acceptance report.
Use this skill to add an AI model provider to LobeHub across runtime integration, model cards, settings, branding, documentation, and acceptance checks. It also covers provider documentation updates and distinguishes new providers from models added to an existing provider.
Use this skill to add a server-side environment variable that sets a default for a user setting. It walks through validation, server configuration and store merging, plus updating the example environment file and self-hosting documentation.
Provides guidance for registering hooks around an agent’s execution steps, tool calls, human approvals, context compression, and sub-agent calls. Hooks can observe events, supply local mock tool results, or use webhooks to allow or deny tool results under configured policies.
Build event-driven background workflows for agents by defining sources, interpreting them as signals, and running actions through policies. It also guides asynchronous handoffs, deduplication, scoped execution, completion receipts, and observability.
Inspect recorded agent runs by operation ID to see failed tool calls, their arguments and results, available tools, execution context, and LLM inputs. The CLI also supports context maps and replaying recorded calls with different models for comparison.