$ loading_
Use AI agents to manage Fugaku jobs, files, and shell commands naturally.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "fugaku-mcp" yet — see the docs or source repo.
Submit an MPI job to Fugaku using 64 nodes for 2 hours, run my_simulation.sh, and return the job ID and submission summary.
A successful submission response with the job ID, resource settings, and command summary.
Check the current status of job 123456 on Fugaku, tell me whether it is queued or running, and summarize the latest log output.
The current job state, queue or runtime details, and a brief summary of key logs.
Download the latest output files from /work/results/ on Fugaku to local storage, and list the transferred files and sizes.
A transfer report with file names, sizes, and any failed items.
Give AI coding agents filesystem, Git, database, and compute tools via MCP.
Connect AI agents to control local JupyterLab for coding and analysis.
Use an AI agent to handle search, files, scraping, and data tasks.
An OAuth-secured remote MCP server for math and Claude Desktop integration.
Provide AI agents with persistent SSH sessions, port forwarding, and file writes.
Automate GitHub issue fixing, testing, PR creation, and production promotion.