Edit videos with OpenClaw

OpenClaw is the self-hosted personal AI agent. Register PandaStudio with one CLI command, then chat with OpenClaw from Slack, Telegram, WhatsApp, Discord, or any of its 20+ connected messaging surfaces to edit videos.

Install PandaStudio + OpenClaw

PandaStudio from writepanda.ai. Launch it once so the local credentials file is written — that's all the setup PandaStudio needs. Then install OpenClaw:

bash

npm install -g openclaw@latest

Run openclaw onboard --install-daemon once to set up the gateway and connect your first chat channel. Verify with openclaw --version.

Enable MCP support inside OpenClaw

OpenClaw ships MCP as an opt-in extra. Install it once:

bash

cd ~/.openclaw/openclaw-agent && uv pip install -e ".[mcp]"

Register PandaStudio with one command

No env var. No config file. No system Node install. PandaStudio ships its MCP server inside the app and exposes it on PATH as pandastudio-mcp — the server auto-discovers credentials at startup.

bash

openclaw mcp add pandastudio --command pandastudio-mcp

On macOS, run sudo /Applications/PandaStudio.app/Contents/Resources/cli/install-cli-symlink-mac.sh once to drop pandastudio-mcp on PATH. The Windows installer handles PATH automatically.

Reload OpenClaw so the new server registers:

bash

openclaw mcp reload

Confirm it started:

bash

openclaw mcp doctor pandastudio --probe

Message OpenClaw to drive PandaStudio

From any of OpenClaw's connected channels (Slack, Telegram, WhatsApp, Discord, the terminal), message OpenClaw:

"Open my latest recording, transcribe it, remove filler words and silences, then export an MP4."
"Create a new 9:16 project, import the file at ~/Recordings/launch.mov, add bold captions, and preview it."
"List all my projects from this week and tell me which one is longest."

Because OpenClaw runs locally and has memory + scheduling, you can also set up recurring jobs ("every Friday, transcribe my new recordings and draft titles for each").

Troubleshooting

Server won't start — run openclaw mcp doctor pandastudio --probe for diagnostic output. Most common cause: PandaStudio hasn't been launched yet, so the local credentials file doesn't exist.
Changes not taking effect — run openclaw mcp reload after any config edit. It refreshes the active MCP runtimes without bouncing the gateway.
"Why no token field?" — PandaStudio's MCP server reads its credentials from the local PandaStudio install (loopback only, never exposed to the internet). The desktop app writes them automatically on first launch. Nothing to copy, nothing to rotate.

Reference

CLI & MCP commands

Browse all 60+ verb-noun commands the agent can call.