The RLM pattern enables processing massive contexts (10M+ tokens) that exceed Claude's context window by recursively chunking, processing, and aggregating results. Instead of failing on large files, u

Estimate the cost of any AI prompt across 19+ models before you run it. Works with OpenAI, Anthropic, Google, DeepSeek, xAI, MiniMax, and local models.

Run an autonomous Open Ralph Wiggum coding loop using OpenCode Zen with free models and automatic fallback.

Use when creating or configuring Claude AI agents using the Agent SDK. Covers agent initialization, configuration, and basic setup patterns.

AI engineering skill for prompt optimization, context inference, and intelligent command routing across different models and use cases

Autonomously draft and polish chapters through multi-skill editorial passes. Use when you have a complete outline and want to produce a polished first draft with iterative refinement.

Invoke COORD_INTEL for investigations, forensics, and intelligence gathering

Invoke COORD_TOOLING for tools and skills development

eld-model-law-discovery

Feature store management skill for online/offline feature serving, feature registration, and training-serving consistency.

Ultrathink LLM-as-Judge validation of completed work. Uses extended thinking by DEFAULT for thorough evaluation.

punderstruck

'Intelligent Git Flow branch creator that analyzes git status/diff and creates appropriate branches following the nvie Git Flow branching model.'

This public intake copy packages `plugins/antigravity-awesome-skills/skills/bdi-mental-states` from `https://github.com/sickn33/antigravity-awesome-skills` into the native Omni Skills editorial shape

This public intake copy packages `plugins/antigravity-awesome-skills-claude/skills/bdi-mental-states` from `https://github.com/sickn33/antigravity-awesome-skills` into the native Omni Skills editorial

Most companies can tell you who their competitors are. Very few can tell you why they're winning or losing against them. Even fewer can tell you whether their position is getting stronger or weaker ri

>

Driver-based budgeting and forecasting skill with rolling forecast support and variance analysis

ASC 606 five-step model implementation skill for revenue recognition analysis and documentation

Analyze skill gaps and prioritize learning investments across the organization

Media interview preparation and crisis simulation tools for executive readiness and spokesperson development

Apply LDA, NMF, and other computational methods to discover patterns in large text corpora with appropriate parameter tuning

This override captures the Codex-native execution model for multi-issue epic work.

This override captures the Codex-native execution model for parallel work.

Analyze the current codebase and produce a structured threat model at `.turbo/threat-model.md`.

>-

Predict project completion dates using ML models. Forecast schedule delays based on current progress, historical patterns, and risk factors.

>

>

Agente que simula Geoffrey Hinton — Godfather of Deep Learning, Prêmio Turing 2018, criador do backpropagation e das Deep Belief Networks.

Sub-skill técnica de Yann LeCun. Cobre CNNs, LeNet, backpropagation, JEPA (I-JEPA, V-JEPA, MC-JEPA), AMI (Advanced Machinery of Intelligence), Self-Supervised Learning (SimCLR, MAE, BYOL), Energy-Based Models (EBMs) e código PyTorch completo.

Traditional physical education rotates through activities on a weekly or biweekly schedule: basketball this week, volleyball next week, fitness the week after. Daryl Siedentop's Sport Education model

Google Ads Scripts Reference — JavaScript automation, AdsApp object model, selectors, GAQL queries, 10+ working script patterns, MCC parallel processing, Sheets integration

explaining-machine-learning-models

Assess bias in medical prediction model studies using PROBAST tool. Use when user wants to evaluate the quality or risk of bias of a medical paper (text or PDF).

Split and read long documents chapter-by-chapter for structured analysis

>

⛁ Manage your statusbar: the persistent gauge showing model, rate limits, context %, directory and git branch at the bottom of Claude Code

> End-to-end threat modeling from system decomposition through threat enumeration, risk rating, and mitigation tracking -- the operational backbone of proactive security design

>

This skill represents the ultimate evolution of web search and AI assistance, integrating **ClaudeCode's revolutionary features** with Manus/Genspark-style multi-model intelligence and ClaudeCowork's

用 OpenClaw 驱动 Codex CLI 的受管运行时。支持交互式 tmux 会话、一次性 exec 任务、会话状态查询、显式 session-id 路由、启动阻塞识别与完成通知。

- **Typed contracts are non-negotiable.** Never widen a contract field to `dict[str, Any]` or `Any`. Every module output is a typed Pydantic `BaseModel`.

- Workspace member: **`rust_scorer`** (CLI + `float_scan_bench`). **`neat-core`** is a **path** dependency on **`../../NEAT-AI-core/neat-core`** (sibling clone of **[NEAT-AI-core](https://github.com/s

An agent in Microsoft Agent Framework is a stateful conversational entity that combines:

Always set `model:` when spawning subagents. Haiku costs 12× less than Sonnet — use it for all search and explore work.

This file defines the lead-agent operating model for Aegis feature work in Claude-driven sessions.

How to use different LLM models for different agents in Present-Agent2.

| Agent ID | Owns | Entry dirs | Model | Can parallel with |

<!--