Execute AI/ML experiments locally or remotely with environment, runtime, and logging controls. Prefer invoking via research-workflow. TRIGGER when: user asks to run/launch/start/resume/monitor a training job, evaluation, or benchmark, or a plan is ready for execution, or experiment needs rerun/recovery. DO NOT TRIGGER when: research investigation only (use deep-research), planning without execution (use research-plan), or env setup without launch (use project-context).

Escalate critical decisions to a human with options, tradeoffs, and recommendation. TRIGGER when: major safety risk, high irreversible impact, large GPU spend, destructive data ops, shared-memory publication, or hard blockers requiring human approval. DO NOT TRIGGER when: routine confirmations handled by run-governor, or low-stakes reversible actions.

Build execution-ready research plans, experiment proposals, and ablation designs. Prefer invoking via research-workflow. TRIGGER when: user asks for research proposal, experiment plan, ablation design, evaluation roadmap, study design, or after deep-research scoping needs conversion to actionable plan. DO NOT TRIGGER when: evidence gathering needed first (use deep-research), actual experiment launch (use experiment-execution), or paper writing (use paper-writing).