Menú
Rules for maintaining research integrity. No simulation, no fabrication, no shortcuts.
Instalar con Codex o Claude Copia este prompt, pégalo en Codex, Claude u otro asistente, y deja que revise la página de la skill y la instale por ti.
Basado en la clasificación ocupacional SOC
Strategies for fitting a paper to a venue's page limit without destroying content quality. Applies to any phase that modifies paper content.
Rules for maintaining figure and table data integrity. All visual data must trace to result files — never invented or modified for convenience.
How to submit SLURM jobs on the SANDS Lab ROCS testbed cluster (KAUST). Covers GPU GRES selection (v100/a100 — p100 is off-policy), QoS caps, conda env activation, multi-node DeepSpeed, preemptible spot jobs, and the watchdog that kills jobs with <15% GPU utilization. Use for any project whose compute_backend is slurm and whose cluster nodes are named `mcnode*`.
Protocol for requesting human intervention when an agent encounters a blocker it cannot resolve autonomously.
Containerized isolation via Apptainer for reproducible experiments. Use whenever the project executes third-party code (agent frameworks, browser agents, codegen tools, benchmark harnesses), runs GPU workloads via Slurm on a shared cluster, needs a reproducible runtime environment for paper claims, or isolates untrusted/adversarial inputs from the host. Broadly applicable to measurement, benchmark, and systems papers — not only to malicious-code experiments.
Best practices for project-isolated environment setup. All installations must be local to the project, never global.
| name | research-integrity |
| description | Rules for maintaining research integrity. No simulation, no fabrication, no shortcuts. |
| tags | ["system","integrity","experiments","ethics"] |
When the research requires evaluating a real system, you MUST use that system.
simulate_*() functions to generate fake metricsnp.random to fabricate resultsWhen the experiment plan requires running a real system (framework, platform, benchmark), you must NOT substitute it with an LLM call:
if not installed → use_llm_instead() branchesBefore declaring any system unavailable:
"install_attempted": false is NEVER acceptable in environment_setup.jsonIf you cannot run an experiment:
After each experiment:
Every quantitative claim in the paper must trace to a real result file:
results/ files% TODO comment — do not invent numbersPapers must comply with double-blind review: