This skill is a “meta-workflow” that sequences existing Numerai skills so research requests reliably produce: (1) runnable configs, (2) executed experiments, (3) a full written report + plots, and (4) a deployable pickle when requested.
Workflow (always follow this order)
1) Design the experiment (use numerai-experiment-design)
Follow the numerai-experiment-design skill to:
clarify the idea (or run quick scout interpretations if ambiguous)
choose baseline + feature set alignment (default ender20 baseline)
create an experiment folder under numerai/agents/experiments/<experiment_name>/
write configs in configs/
run training via PYTHONPATH=numerai python3 -m agents.code.modeling --config <config> --output-dir <experiment_dir>
track metrics with BMC as primary (bmc_mean, bmc_last_200_eras)
iterate in rounds (typically 4–5 configs per round), and keep going until you hit a plateau (per the experiment-design skill)
scale winners (bigger feature set and/or full data) before finalizing the best model
2) Implement new model types if needed (use numerai-model-implementation)
Only if the idea requires new code (new model wrapper, new fit/predict behavior, etc.):
Follow the numerai-model-implementation skill to add the model type and register it.
Add at least one smoke-test config and verify the pipeline runs.
3) Report the research (use report-research)
After you have iterated through multiple rounds and stopped finding improvements (plateau), and after any confirmatory scale runs:
Follow the report-research skill to:
write a full experiment.md (abstract + methods + results + decisions + next steps)
generate the standard show_experiment plot(s)
link plots and artifacts in the report
4) Package and upload (use numerai-model-upload)
If (and only if) the user wants deployment:
Follow the numerai-model-upload skill to create a Numerai-compatible pickle and upload it via the Numerai MCP.
Remember: only Classic (tournament 8) supports pickle uploads.
Defaults (unless user specifies otherwise)
Scout first on downsampled data; scale only winners.
Run experiments in rounds (4–5 configs per round) and stop only after a plateau + confirmatory scale step.
Benchmark reference: v52_lgbm_ender20.
Always record corr + BMC metrics and include the standard plot in the report.
Converted and distributed by TomeVault — claim your Tome and manage your conversions.
1---2name: numerai-example-scripts-numerai-research3description: Numerai Research4---56# Numerai Research78## Overview910This skill is a “meta-workflow” that sequences existing Numerai skills so research requests reliably produce: (1) runnable configs, (2) executed experiments, (3) a full written report + plots, and (4) a deployable pickle when requested.1112## Workflow (always follow this order)1314### 1) Design the experiment (use numerai-experiment-design)1516- Follow the `numerai-experiment-design` skill to:17 - clarify the idea (or run quick scout interpretations if ambiguous)18 - choose baseline + feature set alignment (default ender20 baseline)19 - create an experiment folder under `numerai/agents/experiments/<experiment_name>/`20 - write configs in `configs/`21 - run training via `PYTHONPATH=numerai python3 -m agents.code.modeling --config <config> --output-dir <experiment_dir>`22 - track metrics with BMC as primary (`bmc_mean`, `bmc_last_200_eras`)23 - **iterate in rounds** (typically 4–5 configs per round), and keep going until you hit a plateau (per the experiment-design skill)24 - **scale winners** (bigger feature set and/or full data) before finalizing the best model2526### 2) Implement new model types if needed (use numerai-model-implementation)2728Only if the idea requires new code (new model wrapper, new fit/predict behavior, etc.):29- Follow the `numerai-model-implementation` skill to add the model type and register it.30- Add at least one smoke-test config and verify the pipeline runs.3132### 3) Report the research (use report-research)3334After you have iterated through multiple rounds **and** stopped finding improvements (plateau), and after any confirmatory scale runs:35- Follow the `report-research` skill to:36 - write a full `experiment.md` (abstract + methods + results + decisions + next steps)37 - generate the standard `show_experiment` plot(s)38 - link plots and artifacts in the report3940### 4) Package and upload (use numerai-model-upload)4142If (and only if) the user wants deployment:43- Follow the `numerai-model-upload` skill to create a Numerai-compatible pickle and upload it via the Numerai MCP.44- Remember: only Classic (tournament 8) supports pickle uploads.4546## Defaults (unless user specifies otherwise)4748- Scout first on downsampled data; scale only winners.49- Run experiments in rounds (4–5 configs per round) and stop only after a plateau + confirmatory scale step.50- Benchmark reference: `v52_lgbm_ender20`.51- Always record corr + BMC metrics and include the standard plot in the report.5253---54> Converted and distributed by [TomeVault](https://tomevault.io/claim/numerai) — claim your Tome and manage your conversions.55<!-- tomevault:4.0:skill_md:2026-04-11 -->
Run npx get-skill add tomevault-io/numerai-example-scripts-numerai-research in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.
Numerai Research It is listed under Research & Search on SkillMD.
This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.
This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.
Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.
tomevault-io (@tomevault-io) published this skill. Their other Agent Skills are listed on their SkillMD profile.
