Skip to content
Get packs

Agent Platform Tuning Management

Lets an agent list, check, and (with typed confirmation) cancel Google Cloud GenAI model tuning jobs

Source Download skill
Repository
google/skills
Stars
21,782
Licence
Apache-2.0
Files
1 file

What it does

This skill teaches an AI agent how to check on model tuning jobs (customizing an AI model with extra training data) in Google Cloud's Agent Platform. The agent can list your tuning jobs, look up the status of one job by ID, and cancel a running job. Listing and checking happen right away, while cancelling requires you to show a preview and type an explicit confirmation first.

Great for

  • Engineers who run model tuning jobs on Google Cloud and want status updates from their agent
  • Teams that need to find a stuck or runaway tuning job and stop it safely

Skip it if

  • Starting new fine-tuning runs; the file says to use the separate agent-platform-tuning skill for that
  • Deploying models or managing serving endpoints, which are covered by other skills; cancelling a job also halts any in-progress training permanently, so only use it when you are sure

Try asking

  • “What tuning jobs do I have running in project my-project in us-central1?”
  • “Check the status of tuning job 1234567890123456789.”
  • “Cancel tuning job 1234567890123456789 that has been running too long.”
Step-by-step guide: how to use Agent Platform Tuning ManagementInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 1 original files, the Apache-2.0 licence and an attribution note.

    agent-platform-tuning-management.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip agent-platform-tuning-management.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip agent-platform-tuning-management.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Give the agent your project ID and region up front so it does not have to ask.
  • When cancelling, read the preview card carefully to confirm the job ID before typing 'I confirm'.
  • Ensure you are authenticated to the correct Google Cloud project before running commands, since the agent will act on whatever credentials are active.

What's inside

  • SKILL.md

Redistributed unmodified from google/skills under the Apache-2.0 licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.