jetbrains

Watch Skill in JetBrains IDEs

Status: doc-verified ☑ — config matches JetBrains' own Junie and AI Assistant MCP documentation; not executed here.

Covers Junie (IDE plugin and CLI) and AI Assistant across IntelliJ IDEA, PyCharm, WebStorm, GoLand, and the rest of the family.

Install

uvx --from "watch-skill[standard]" watch-skill doctor

uvx fetches the package on first use and needs no checkout. Prefer a permanent install? pipx install "watch-skill[standard]", then use "command": "watch-skill", "args": ["serve"] below.

git clone https://github.com/oxbshw/watch-skill && cd watch-skill
uv sync --extra all
uv run watch-skill doctor

Configure Junie

Junie reads the same JSON in the IDE plugin and the CLI. Project scope is .junie/mcp/mcp.json in the repository root; global scope is ~/.junie/mcp/mcp.json, which is also what the Tools → Junie → MCP Settings page writes.

{
  "mcpServers": {
    "watch-skill": {
      "command": "uvx",
      "args": ["--from", "watch-skill[standard]", "watch-skill", "serve"]
    }
  }
}

Commit the project file to give a team the same video tools.

Configure AI Assistant

Settings → Tools → AI Assistant → Model Context Protocol → Add, then paste the same block. Local servers use command and args; the url form is for remote servers, which is what watch-skill serve --http would be.

Smoke test (3 steps)

1. Ask: "Watch https://www.youtube.com/watch?v=aqz-KE-bpKQ and tell me what happens at 0:10."
2. Approve the watch_video call.
3. Follow up: "what color is the bird?" — should call ask_video and answer from the index without re-processing.

Notes

• Junie and AI Assistant keep separate MCP configuration. Setting one up does not configure the other.
• A remote setup is watch-skill serve --http (port 8747, /mcp), entered as a url rather than a command. Keep it on loopback or put a bearer token in front of it — see Configuration.