meta AI Agent Skills

Browse 30 skills related to meta

llamaguard

Meta's 7-8B specialized moderation model for LLM input/output filtering. 6 safety categories - violence/hate, sexual content, weapons, substances, self-harm, criminal planning. 94-95% accuracy. Deploy with vLLM, HuggingFace, Sagemaker. Integrates with NeMo Guardrails.

Safety AlignmentLlamaGuardContent Moderation+6

torchforge-rl-training

Provides guidance for PyTorch-native agentic RL using torchforge, Meta's library separating infrastructure from algorithms. Use when you want clean RL abstractions, easy algorithm experimentation, or scalable training with Monarch and TorchTitan.

Reinforcement LearningPyTorchGRPO+4

capability-evolver

A self-evolution engine for AI agents. Analyzes runtime history to identify improvements and applies protocol-constrained evolution.

metaaiself-improvement+1

skill-creator

Joelhooks Swarm Tools Skill Creator

Guide for creating effective agent skills. Use when you want to create a new skill, improve an existing skill, or learn best practices for skill development. Helps codify learned patterns into reusable, discoverable skills.

metaskillslearning+1

find-skills

Discover and install skills from multiple marketplaces for AI coding agents

metadiscoverymarketplace+1

Skill Authoring

Guidelines for writing effective AI agent skills

metaauthoringskills+1

skill-creator

Create new reusable skills from conversation context

metaproductivity

help

OrchestKit skill directory with categorized listings. Use when discovering skills for a task, finding the right workflow, or browsing capabilities.

helpdocumentationskills+2

meta

Metacognition skill for AI agents. Use when starting work, feeling stuck, output feels off, or before complex tasks. Teaches how to think about thinking.

metametacognition

simulator-effect

Implication beats simulation — imagination renders (Will Wright)

moollmmetaphilosophy+4

society-of-mind

Intelligence emerges from many simple agents (Minsky)

moollmmetaphilosophy+4

bootstrap

Wake up, orient, and warm the context with foundational MOOLLM knowledge

moollmmetaorientation+3

coherence-engine

The LLM as consistency maintainer and orchestrator

moollmmetaorchestration+2

skill

A skill is documentation that learned to do things.

moollmmetaprogram+2

leela-ai

Manufacturing Intelligence — Leela AI develops MOOLLM for practical use and industrial exploration

moollmmetacompany+4

manufacturing-intelligence

The slogan unpacked — seven readings of 'Manufacturing Intelligence'

moollmmetaphilosophy+5

moollm

The soul of MOOLLM — self-explanation, help, navigation, philosophy

moollmmetahelp+3

quality-audit

Meta-skill for auditing and validating skill quality. Use when reviewing skills for consistency, completeness, accuracy, and adherence to standards. Provides structured rubrics, scoring frameworks, and actionable recommendations.

metaqualityvalidation+2

skill-harvester

Meta-skill for extracting and creating reusable Claude Code skills from past work sessions. Analyzes git history, code patterns, workflows, and documentation to identify harvestable skills, then generates comprehensive skill definitions with best practices, examples, and structured templates.

metaautomationskill-creation+2

skill-chain-prompts

organvm-iv-taxis

Orchestrate multi-skill workflows with prompt-based skill chaining. Define sequential or parallel skill invocations using YAML chain definitions, track progress through chains, and use pre-built chains for development, documentation, and career workflows. Use when coordinating multiple skills for complex tasks.

orchestrationworkflowchaining+3

generate-skill

Use when asked to "create a skill", "generate a SKILL.md", "make me a skill", "build a custom skill", or when user wants to extend Claude Code capabilities with a new skill

skill-creationmetaautomation+1

rl-innovator

Meta-skill that orchestrates other RL skills to explore novel algorithm designs, identify limitations, and propose mathematically grounded improvements.

orchestrationinnovationmeta

create-skill

Wizard for creating new Agent Zero skills. Guides users through creating well-structured SKILL.md files. Use when users want to create custom skills.

metawizardcreation+2

capability-evolver

A self-evolution engine for AI agents. Analyzes runtime history to identify improvements and applies protocol-constrained evolution.

metaaiself-improvement+1

code-first-planner

Generate implementation plans programmatically using Python instead of writing markdown manually. Produces consistent phase files with YAML frontmatter, cross-references, validation scripts, and checkpoint support (.toon format). Use when creating any multi-phase plan, project scaffold, or structured documentation.

planningcode-firstmeta+2

skill-master

Meta-skill for creating and managing Claude Code skills. Reference for SKILL.md frontmatter, progressive disclosure tiers, skill types, naming conventions, and validation checklists.

metaskillsinfrastructure+1

hook-master

Meta-skill for creating, validating, testing, and debugging Claude Code hooks. Comprehensive reference for all 18 hook events, exit code contracts, JSON schemas, and templates for both Python and shell hooks.

metahooksinfrastructure+1

skill-writer

Concise 10-step workflow for creating Claude Code skills. Quick-reference guide that gets you from idea to validated skill in minimal steps.

metaskillsworkflow+1

continuous-learning-v2

Instinct-based pattern learning with confidence scoring and auto-evolution

learningpatternsinstincts+2

agent-writer

Write new agent definitions, skill files, and agent configurations. Use when someone asks to "create a new agent", "write a skill", "define an agent", or needs to create reusable AI agent instructions for any platform. This is the meta-skill for creating other skills and agents in the polyagent-skills format.

agentskillmeta+2