trl-training
Train and fine-tune transformer language models using TRL (Transformers Reinforcement Learning). Supports SFT, DPO, GRPO, KTO, RLOO and Reward Model training…
Skills
Find an Agent Skill for your task. Read the guide and copy the install command.
Updated 10-28 00:17
Agents & models
24,984 results
memory
Give agents memory that survives the conversation — facts recalled by meaning (ctx.memory.add / search) and JSON documents recalled by key (collections),…
thrifty
Use when the user says "thrifty", "delegate this", "tiered build", or asks to execute a spec cheaply. thrifty takes a spec you already have (written by hand or…
bedrock-agentcore-evaluations
Amazon Bedrock AgentCore Evaluations for testing and monitoring AI agent quality. 13 built-in evaluators plus custom LLM-as-Judge patterns. Use when testing…
esm
ESM protein language models for embeddings, sequence scoring, structure prediction, and binder design. Use this skill when: (1) Computing pseudo-log-likelihood…
rfdiffusion
Generate protein backbones using RFdiffusion, a diffusion-based generative model for de novo protein structure generation. Use this skill when: (1) Designing…
codex-agent-loop
Understand and apply how the Codex agent loop works, including request assembly (`instructions`, `tools`, `input`), thread/turn/item mental models, tool…
find-skills
Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express…
skill-creator
Creates, updates, routes, adopts, promotes, imports, validates, registers, and distributes agent skills through the `/Users/dobby/GitHub/agents` control-plane…
skill-router
Deprecated. Skill routing has been absorbed by `skill-creator`. Use `skill-creator` for deciding `owned`, `external`, `repo-local`, or `dormant` placement;…
whiyh-gpt-hill-climb
Monitor, analyze, and improve the Who's In Your Head GPT/default game path. Use when Codex is asked to check recent games, misses, dropoffs, abandoned…
vuln-defectdojo
Vulnerability management and findings aggregation using DefectDojo. Centralizes security findings from all SecOpsAgentKit scanners (Semgrep, Bandit, ZAP,…
arc-parallel-implement
Implements multiple planned user stories in parallel with isolated worktrees, using Arc Pi native subagents when available or the non-Pi external orchestrator…
wan-ascend-adaptation
This skill provides comprehensive guidance for adapting Wan-series video generation models (Wan2.1/Wan2.2) from NVIDIA CUDA to Huawei Ascend NPU. It should be…
grad-signaling
Apply signaling theory (Spence, 1973) to analyze how agents communicate private information through costly, credible signals under information asymmetry. Use…
crow-reason
Critical reasoning and devil's advocate. Steelmans your position, then finds the cracks. Use when you need assumption-challenging, pre-mortems, red-teaming, or…
groundhog-surface
Surface and validate assumptions before committing to action. Pop up, look around, report what's real and what's assumed, then persist the validated ground for…
strands-context-manager
Strands conversation/context manager patterns, including sliding window with summarization. Use when building agents with context management, preventing…
neo-sub-agent
Use this skill when the user asks to design, create, review, or convert sub-agents, custom agents, worker/reviewer/planner agents, background agents, or…
jb-beansloop
Use when working one bean or all beans in a beans-enabled repo through the Ralp loop.
turn-into-app
Turns a thread, skill, spreadsheet, or Claude/ChatGPT project into a polished, visual Agent-Native app: populated domain screens with the in-app agent working…
agent-first-repo
Structure a repository and its documentation so AI coding agents can work effectively. Covers knowledge hierarchy, progressive disclosure, mechanical…
wren-usage
Wren Engine CLI workflow guide for AI agents. Answer data questions end-to-end using the wren CLI: gather schema context, recall past queries, write SQL…
find-skills
Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express…