Description
run-train is an agent skill from the rigorpilot-skills collection for running a training command conservatively. It is the Rigor Train skill, and its installed name is kept for compatibility. It is used once a training command has already been selected, whether it was documented in the repository or chosen by the researcher, and it keeps the training evidence bounded while leaving repository-specific monitoring details to the model. The skill covers startup verification, a short run to check that training works, the full training kickoff and the handling of resumed runs. It is built to report training in a structured way, with status, checkpoints and metrics, so that a researcher can see what happened and where a run stands. Because training can take hours and use a lot of compute, the skill favors a careful start, with a verification step before a long run is allowed to go on. The skill points to the shared operating principles for the research repository before it works. It should not be used to choose a training command or to plan experiments, which belong to the explore and reproduction skills. It is a practical choice for a researcher who wants an agent to launch and supervise a training job with clear checkpoints. The skill passes Socket and Snyk, and Gen Agent Trust Hub shows a warning on skills.sh. Install it with npx skills add https://github.com/lllllllama/rigorpilot-skills --skill run-train.…
Highlights
- Runs a selected training command conservatively
- Startup verification and a short run before the full kickoff
- Handles resumed training runs
- Reports status, checkpoints and metrics in a structured way
Ratings & Reviews
No ratings yet
Be the first to rate run-train.
People also visit
run-train FAQ
When should I use it?
When a training command has been selected and should be run carefully, from a startup check to a full run or a resume.
Does it choose the command?
No. It runs a command that was already documented or selected, and command selection belongs to the reproduction skill.
What does it report?
Structured training status, checkpoints and metrics.
How do I install it?
Run npx skills add https://github.com/lllllllama/rigorpilot-skills --skill run-train.
Which security audits does it have?
On skills.sh, Gen Agent Trust Hub shows a warning, and Socket and Snyk pass.
What are the best run-train alternatives?
Popular alternatives to run-train include ai-research-explore, analyze-project and ai-research-reproduction. See all run-train alternatives