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 K-Dense-AI/mimeo --skill david-silvergit clone --depth 1 https://github.com/K-Dense-AI/mimeoWrote 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/k-dense-ai/mimeo/david-silver)<a href="https://agentmods.dev/skills/k-dense-ai/mimeo/david-silver"><img src="https://agentmods.dev/badge/skills/k-dense-ai/mimeo/david-silver/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/k-dense-ai/mimeo/david-silver"><img src="https://agentmods.dev/badge/skills/k-dense-ai/mimeo/david-silver.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 | $0.00127 | $0.01252 |
| Opus 5.5 | $0.00051 | $0.00501 |
| Sonnet 5.5 | $0.00025 | $0.00250 |
| Haiku 4.5 | $0.00013 | $0.00125 |
Grade A, and why
david-silver 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.
Copies of this mod
1 near-identical copy found in the catalogue:
- david-silver — 100% identical, 2 lines differ
How it starts
The opening of the file, as written. The whole thing — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Thinking like David Silver
David Silver is a pioneering reinforcement learning researcher and the lead researcher on AlphaGo and AlphaZero at DeepMind. His signature thinking style revolves around the conviction that true intelligence emerges not from mimicking human data, but from autonomous trial-and-error learning. He views intelligence as a formalizable reinforcement learning problem where agents interact with an environment to maximize expected cumulative reward.
His approach fundamentally rejects the "knowledge acquisition bottleneck"—the idea that we must hand-code human heuristics into machines. Instead, he advocates for tabula rasa (blank slate) learning, where systems discover novel, superhuman strategies purely through self-play and experience.
Reach for this skill whenever you're designing AI training loops, evaluating the limits of human data (like LLMs), balancing exploration and exploitation, or selecting ambitious research problems in machine learning.
Core principles
- The Era of Experience Over Human Data: Human data bootstraps learning but caps performance at human levels; superhuman intelligence requires continuous learning from the agent's own experience.
- Tabula Rasa Learning Surpasses Human Expertise: Pure reinforcement learning without human knowledge or domain-specific tuning scales further and discovers superior, counterintuitive solutions.
- The Purity of Self-Learning: Hardcoding human heuristics fits the algorithm to human biases; throwing out human data forces the creation of infinitely scalable self-learning mechanisms.
- The Reward Hypothesis: All goals can be formalized as the maximization of expected cumulative reward, providing a single axis to evaluate conflicting objectives.
For detailed rationale and quotes, see references/principles.md.
How David Silver reasons
Silver approaches AI development by looking for "microcosms"—environments with simple rules but vast emergent complexity (like Go or chess) that allow for rapid iteration without the friction of the physical world. When evaluating a system, he asks whether it is merely distilling existing knowledge (the "shallow problem") or learning to discover new knowledge for itself (the "deep problem").
What ships with it
9 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 Changed · +2 lines 9a10e92cfc78
- 1mo ago First seen · 65 lines · 127 tokens per session scan A a2aca8e88640
david-silver is a skill published in the GitHub repository K-Dense-AI/mimeo (285 stars, last pushed 1mo ago), licensed MIT. It adds 127 tokens to every session and 1,252 once invoked, about $0.0005 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 2026-08-30.
Other skills, from other repositories
arboreto
Infers candidate gene regulatory networks from bulk or single-cell expression data using AertsLab Arboreto GRNBoost2 and GENIE3. Use for transcription factor-target association ranking, compatible Dask execution, sparse expression inputs, and network stability checks.
deepspot-m
Generates transcriptome-wide virtual spatial transcriptomics from H&E histology with DeepSpot-M. Used for predicted log1p-CPM expression from 224x224 tiles at about 20x, querying the released protein-coding gene panel by symbol, and whole-slide prediction after resolution-aware tiling with histolab.
torchdrug
Builds and troubleshoots TorchDrug 0.2.1 workflows for molecular graphs, property prediction, self-supervised pretraining, molecule generation, retrosynthesis, protein representation learning, and knowledge graph reasoning. Use when code imports torchdrug or needs its datasets, models, tasks, or Engine.
pyhealth
Builds and validates PyHealth clinical machine-learning pipelines for EHR, signals, imaging, and medical codes. Use for PyHealth dataset loading, MIMIC-III/IV, eICU or OMOP prediction tasks, patient-level evaluation, mortality/readmission/length-of-stay modeling, medication recommendation, sleep staging, Trainer…
ml-training-recipes
Battle-tested PyTorch training recipes for all domains — LLMs, vision, diffusion, medical imaging, protein/drug discovery, spatial omics, genomics. Covers training loops, optimizer selection (AdamW, Muon), LR scheduling, mixed precision, debugging, and systematic experimentation. Use when training or fine-tuning…
nemo-mbridge-perf-expert-parallel-overlap
Validate and use MoE expert-parallel communication overlap in Megatron-Bridge, including overlapmoeexpertparallelcomm, delaywgradcompute, and flex dispatcher backends such as DeepEP and HybridEP.