Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add topprismdata/cultivating-ml-agent --skill competition-orchestration-multimodelgit clone --depth 1 https://github.com/topprismdata/cultivating-ml-agentWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/topprismdata/cultivating-ml-agent/competition-orchestration-multimodel)<a href="https://agentmods.dev/skills/topprismdata/cultivating-ml-agent/competition-orchestration-multimodel"><img src="https://agentmods.dev/badge/skills/topprismdata/cultivating-ml-agent/competition-orchestration-multimodel/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/topprismdata/cultivating-ml-agent/competition-orchestration-multimodel"><img src="https://agentmods.dev/badge/skills/topprismdata/cultivating-ml-agent/competition-orchestration-multimodel.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00226 | $0.02579 |
| Opus 5 | $0.00113 | $0.01290 |
| Sonnet 5 | $0.00045 | $0.00516 |
| Haiku 4.5 | $0.00023 | $0.00258 |
Grade A, and why
competition-orchestration-multimodel scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 221 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Competition Orchestration: Fork Public Submissions to Maximize LB
Problem
Kaggle competitions often have:
- Strong public kernels scoring close to top LB (e.g., NeuroGolf baseline 7228 → forked lucifer 7269.68)
- Re-runs take 10-15 hours per submission (BUDGET-FILLING attacks, full test inference)
- Limited quota: 5 submissions/day max, but each takes hours to score
- Top scorers use LLM agent loops that may be inaccessible to a solo agent
A solo agent (no GPU, no API budget) can still achieve top-25% by:
- Forking the highest-scoring public kernels instead of building from scratch
- Running many parallel submissions of different approaches
- Knowing which public datasets to import as kernel inputs
Context / Trigger Conditions
Use this skill when:
- A public dataset contains
submission.zipwith the exact format you need - Top kernels are public and within reach of your submission
- A competition allows re-using public kernels as input
- You see "0.0" or "ERROR" submissions that don't seem like real errors
- Time pressure: deadline in <2 weeks
- A competition's hidden test set rewards public knowledge
Solution
Lesson 1: Fork Public Submissions Aggressively (NeuroGolf Case)
Concrete case (2026-07-13): NeuroGolf baseline scored 7228. Within 4 hours, forked 7 public kernels via inline-base64, raising score to 7269.68:
| Submission | Score | Source |
|---|---|---|
| baseline_7238 (rescue) | 7228.04 | self-fork |
| franksunp 7267.99 | 7267.99 | kernel pull, extract B64 |
| uditjain 7268.48 | 7268.48 | dataset source via dataset_sources |
| lucifer 7269.64 | 7269.68 | kernel pull, base64 decode |
| kojimar 7169.36 | 7169.36 | dataset + read_task_zip blend |
| octaviograu 6154.71 | 6154.71 | dataset source path discovery |
| boristown V176 | 7266.73 | kernel pull, base64 decode |
Pattern: Each kernels pull + base64 decode + write submission.csv is one PR.
Most public kernels embed submission as base64 (1-2MB). For dataset-based:
mount path is /kaggle/input/datasets/<owner>/<name>/submission/.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 11d ago First seen · 221 lines · 226 tokens per session scan A e09ebdcf4c7e
competition-orchestration-multimodel is a skill published in the GitHub repository topprismdata/cultivating-ml-agent (5 stars, last pushed 14d ago), licensed MIT. It adds 226 tokens to every session and 2,579 once invoked, about $0.0011 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
huggingface-hub
Hugging Face Hub CLI (hf) — search, download, and upload models and datasets, manage repos, query datasets with SQL, deploy inference endpoints, manage Spaces and buckets.
writing
A writing guide for turning verified facts and calculations into finished text for a specific audience. It follows the requested language, structure, and length.
regex-mastery
Use this skill when writing regular expressions, debugging pattern matching,optimizing regex performance, or implementing text validation. Triggers on regex, regular expressions, pattern matching, lookahead, lookbehind, named groups, capture groups, backreferences, and any task requiring text pattern matching.
ws-ckpt
A Linux-only tool for creating and managing workspace snapshots. A snapshot is a saved view of files that can later be compared, restored, listed, or deleted.
food-order
Reorder previous Foodora orders, preview cart contents, and track delivery ETA/status with ordercli. Use when the user wants to reorder food, check delivery status, or browse recent Foodora order history. Never confirm an order without explicit user approval.
mnemon
Persistent memory for MiniMax Code. Recall durable context, store important facts and decisions, and link related memories with the mnemon CLI.