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 unified-vs-day-specific-forecastinggit 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/unified-vs-day-specific-forecasting)<a href="https://agentmods.dev/skills/topprismdata/cultivating-ml-agent/unified-vs-day-specific-forecasting"><img src="https://agentmods.dev/badge/skills/topprismdata/cultivating-ml-agent/unified-vs-day-specific-forecasting/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/unified-vs-day-specific-forecasting"><img src="https://agentmods.dev/badge/skills/topprismdata/cultivating-ml-agent/unified-vs-day-specific-forecasting.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 | $0.00101 | $0.00728 |
| Opus 5 | $0.00051 | $0.00364 |
| Sonnet 5 | $0.00020 | $0.00146 |
| Haiku 4.5 | $0.00010 | $0.00073 |
Grade A, and why
unified-vs-day-specific-forecasting 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 7d 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 — 75 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Unified Model vs Day-Specific in Multi-Step Forecasting
The Assumption
Day-specific models (one per horizon day) should outperform a single unified model because:
- Each model specializes in its specific horizon
- No need for the model to "figure out" how day offset affects predictions
- Each model has its full capacity for one task
The Finding
In Kaggle Favorita Store Sales (16-day horizon, 1782 store-family pairs):
| Approach | CV RMSLE | LB |
|---|---|---|
| Day-specific (16 models) | 0.42567 | 0.39779 |
| Unified (1 model + target_day_offset) | 0.38206 | 0.38850 |
| Blend 70/30 | — | 0.39393 |
| Blend 50/50 | — | ~0.391 |
Unified model is 0.00929 LB better than day-specific.
Why Unified Wins Here
-
More training data: Unified sees 16× more samples (46M vs 2.87M per model), giving LightGBM more statistical power to learn rare patterns.
-
Cross-day generalization: The unified model learns that "sales patterns on Monday" are similar whether predicting day 1 (if target is Monday) or day 8 (if target is Monday). Day-specific models can't share this knowledge.
-
Rich features encode temporal structure: With YoY, TE, and lag features, the model already has enough information to distinguish between horizons. The
target_day_offsetfeature is sufficient for the model to specialize internally. -
Day-specific overfits the fold structure: Each day-specific model trains on the same store-family pairs with very similar time splits, potentially overfitting to the CV fold boundaries.
When Day-Specific Might Still Win
- Very long horizons (>30 days) where temporal patterns change dramatically
- When features are minimal (model needs explicit specialization)
- When different horizons have fundamentally different data availability
Rule of Thumb
- Rich features + moderate horizon (7-30 days): Try unified first
- Sparse features + long horizon: Day-specific may be better
- Always test both and use controlled comparison (not just CV)
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.
- 7d ago First seen · 75 lines · 101 tokens per session scan A 987df220af0a
unified-vs-day-specific-forecasting is a skill published in the GitHub repository topprismdata/cultivating-ml-agent (5 stars, last pushed 14d ago), licensed MIT. It adds 101 tokens to every session and 728 once invoked, about $0.0005 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 2026-09-03.
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.