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 itsmostafa/llm-engineering-skills --skill rlhfgit clone --depth 1 https://github.com/itsmostafa/llm-engineering-skillsWrote 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/itsmostafa/llm-engineering-skills/rlhf)<a href="https://agentmods.dev/skills/itsmostafa/llm-engineering-skills/rlhf"><img src="https://agentmods.dev/badge/skills/itsmostafa/llm-engineering-skills/rlhf.svg" alt="Measured on agentmods" 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 | $0.00040 | $0.02957 |
| Opus 5 | $0.00020 | $0.01478 |
| Sonnet 5 | $0.00008 | $0.00591 |
| Haiku 4.5 | $0.00004 | $0.00296 |
Grade A, and why
rlhf 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 8d 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 — 365 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Understanding RLHF
Reinforcement Learning from Human Feedback (RLHF) is a technique for aligning language models with human preferences. Rather than relying solely on next-token prediction, RLHF uses human judgment to guide model behavior toward helpful, harmless, and honest outputs.
Table of Contents
- Core Concepts
- The RLHF Pipeline
- Preference Data
- Instruction Tuning
- Reward Modeling
- Policy Optimization
- Direct Alignment Algorithms
- Challenges
- Best Practices
- References
Core Concepts
Why RLHF?
Pretraining produces models that predict likely text, not necessarily good text. A model trained on internet data learns to complete text in ways that reflect its training distribution—including toxic, unhelpful, or dishonest patterns. RLHF addresses this gap by optimizing for human preferences rather than likelihood.
The core insight: humans can often recognize good outputs more easily than they can specify what makes an output good. RLHF exploits this by collecting human judgments and using them to shape model behavior.
The Alignment Problem
Language models face several alignment challenges:
- Helpfulness: Following instructions and providing useful information
- Harmlessness: Avoiding toxic, dangerous, or inappropriate outputs
- Honesty: Acknowledging uncertainty and avoiding fabrication
- Intent alignment: Understanding what users actually want, not just what they say
RLHF provides a framework for encoding these properties through preference data.
Key Components
- Preference data: Human judgments comparing model outputs
- Reward model: A learned function approximating human preferences
- Policy optimization: RL algorithms that maximize expected reward
- Regularization: Constraints preventing deviation from the base model
What ships with it
3 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.
- 8d ago First seen · 365 lines · 40 tokens per session scan A 06ca7de506f6
rlhf is a skill published in the GitHub repository itsmostafa/llm-engineering-skills (23 stars, last pushed 4mo ago), licensed MIT. It adds 40 tokens to every session and 2,957 once invoked, about $0.0002 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 2026-08-30.
Other skills, from other repositories
kaggle-learner
This skill should be used when the user asks to "learn from Kaggle", "study Kaggle solutions", "analyze Kaggle competitions", or mentions Kaggle competition URLs. Provides access to extracted knowledge from winning Kaggle solutions across NLP, CV, time series, tabular, and multimodal domains.
llm-as-computer
Execute programs on a compiled transformer stack machine where every instruction fetch and memory read is a parabolic attention head. Demonstrates that transformer attention + FF layers can implement a working computer. Use when user mentions "llm-as-computer", "lac", "stack machine", "compiled transformer"…
better-prompt
A prompt editor that turns rough instructions for AI systems into clearer, more complete prompts. It follows published OpenAI and Anthropic guidance.
ai-engineer-expert
Expert-level AI implementation, deployment, LLM integration, and production AI systems. Use when the user mentions AI engineering, LLM, deployment, production AI, or integration, or when the task involves LLM Patterns, LLM Integration, or Production Systems.
bio-prefect-dask-nextflow
Design reproducible bioinformatics pipelines with Prefect plus Dask or Nextflow. Use when scaffolding local, distributed, or scheduler-backed workflows.
claude-cookbooks
Claude AI cookbooks - code examples, tutorials, and best practices for using Claude API. Use when learning Claude API integration, building Claude-powered applications, or exploring Claude capabilities.