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 nirholas/three.ws --skill recommendation-algorithm-designgit clone --depth 1 https://github.com/nirholas/three.wsWrote 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/nirholas/three.ws/recommendation-algorithm-design)<a href="https://agentmods.dev/skills/nirholas/three.ws/recommendation-algorithm-design"><img src="https://agentmods.dev/badge/skills/nirholas/three.ws/recommendation-algorithm-design/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/nirholas/three.ws/recommendation-algorithm-design"><img src="https://agentmods.dev/badge/skills/nirholas/three.ws/recommendation-algorithm-design.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00088 | $0.04492 |
| Opus 5.5 | $0.00035 | $0.01797 |
| Sonnet 5.5 | $0.00018 | $0.00898 |
| Haiku 4.5 | $0.00009 | $0.00449 |
Grade A, and why
recommendation-algorithm-design 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 1mo 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 — 513 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Recommendation Algorithm Design — From First Principles
This skill teaches you to build recommendation systems that surface the right content to the right person at the right time. You'll learn the full pipeline — from candidate generation to final ranking — and understand why feeds feel the way they do.
The Recommendation Pipeline
Every major social platform follows the same high-level architecture:
┌─────────────────────────────────────────────────────────────┐
│ THE RECOMMENDATION PIPELINE │
├─────────────────────────────────────────────────────────────┤
│ │
│ ALL CONTENT (millions) │
│ │ │
│ ▼ │
│ ┌─────────────┐ │
│ │ CANDIDATE │ Filter: ~10,000 posts worth considering │
│ │ GENERATION │ (follows, topics, graph neighbors) │
│ └──────┬──────┘ │
│ │ │
│ ▼ │
│ ┌─────────────┐ │
│ │ RANKING │ Score each candidate: P(engagement) │
│ │ MODEL │ (neural network, 100+ features) │
│ └──────┬──────┘ │
│ │ │
│ ▼ │
│ ┌─────────────┐ │
│ │ FILTERING │ Remove: spam, duplicates, policy │
│ │ & MIXING │ Inject: diversity, exploration, ads │
│ └──────┬──────┘ │
│ │ │
│ ▼ │
│ FINAL FEED (~50 posts for this session) │
│ │
└─────────────────────────────────────────────────────────────┘
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.
- 1mo ago First seen · 513 lines · 88 tokens per session scan A d02f10d9edbe
recommendation-algorithm-design is a skill published in the GitHub repository nirholas/three.ws (241 stars, last pushed yesterday), licensed Apache-2.0. It adds 88 tokens to every session and 4,492 once invoked, about $0.0004 per session on Opus 5.5. 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
llava
Vision-language chat: VQA, captioning, image dialogue.
sglang-diffusion-cache-dit
Workflow for upgrading/integrating cache-dit in SGLang diffusion (multimodalgen): DBCache, DMD calibrator, TaylorSeer, SVDQuant DQ; porting upstream PRs and resolving conflicts against the per-request knob system; adding new cache knobs; building the sglang generate CLI test matrix; precision validation (PSNR / log…
cellxgene-census
Query the CELLxGENE Census (61M+ cells) programmatically. Use when you need expression data across tissues, diseases, or cell types from the largest curated single-cell atlas. Best for population-scale queries, reference atlas comparisons. For analyzing your own data use scanpy or scvi-tools.
markitdown
Convert files and office documents to Markdown. Supports PDF, DOCX, PPTX, XLSX, images (with OCR), audio (with transcription), HTML, CSV, JSON, XML, ZIP, YouTube URLs, EPubs and more.
flux2-lora-training
Plan or review LoRA and edit-training work specifically for FLUX.2 Klein or Qwen-Image-Edit, including paired datasets, trainer-version contracts, and held-out fidelity checks. Do not use for generic Stable Diffusion/DiT training, prompt authoring, or model serving; route those tasks to their specialized skill.
liter-llm-calling-llms
Use when sending chat completions through liter-llm and routing to a specific provider via the provider/model prefix. Covers the chat call shape, provider routing, modelhint, message roles, and error categories.