| name | submit-experiment-results |
| description | Submit experiment results for advisor review. Commits changes, pushes the branch, marks the PR as ready, and swaps the status label from wip to review. Use this skill when you've finished running experiments and posted your results comment. Triggers for: "submit for review", "mark PR ready", "send results to advisor", "submit experiment results".
|
| argument-hint | <pr-number> <problem-dir> |
| model | claude-sonnet-4-6 |
| effort | high |
submit-experiment-results
You've run the experiment, posted a results comment on the experiment PR — now wrap it up and hand it to the advisor.
Arguments
- $0 — The experiment PR number (e.g.
1842)
- $1 — The active problem directory (e.g.
target/cfd_tandemfoil)
Before you call this
Make sure you've already posted a results comment on the experiment PR with metrics, W&B run ID, analysis, and suggested follow-ups.
The comment must include a valid one-line terminal result marker before the Results section:
SENPAI-RESULT: {"terminal":true,"status":"complete","pending_arms":false,"wandb_run_ids":["<run-id>"],"primary_metric":{"name":"<metric>","value":<number>},"test_metric":{"name":"<metric>","value":<number>}}
Use terminal=true only when every advisor-required arm/run is finished or intentionally aborted. If any arm or W&B run can still change the conclusion, leave the PR as status:wip.
Steps
- Stage and commit your changes:
git add "$1/train.py"
git commit -m "<concise description of what you changed>"
- Push the branch:
git push origin "$(git branch --show-current)"
- Mark the PR as ready for advisor review and swap labels:
source "${CLAUDE_PLUGIN_ROOT}/scripts/senpai-gh.sh"
mark_ready_for_review $0
mark_ready_for_review refuses the handoff if the terminal SENPAI-RESULT marker is missing, invalid JSON, or still reports pending arms/runs. Read the error, fix the PR comment, and retry.
That's it. The advisor will pick it up in their next review cycle.
