Agent Skills Hub

debug-distributed

Debugging guide for AReaL distributed training issues, including hangs, NCCL errors, OOM, and numerical consistency in FSDP2/TP/CP/EP.

constant-time-testing

Detects timing side channels in cryptographic code to prevent secret data leakage. Essential for auditing sensitive implementations.

trulens-evaluation-workflow

A systematic workflow to instrument, evaluate, and monitor LLM applications using TruLens, supporting frameworks like LangChain, LangGraph, and LlamaIndex.

sound-effects

Generate professional-grade sound effects from text descriptions. Create audio textures, cinematic impacts, UI sounds, and ambient environments with precision control over duration, looping, and prompt adherence.

prompt-writing

Create, refine, and optimize high-quality YAML prompts for AI assistants using structure guidelines, template patterns, and quality standards.

marketing-ideas

A strategic marketing ideation engine for SaaS founders. Generate actionable, stage-appropriate growth strategies, content tactics, and promotional ideas tailored to your budget and specific product context.

tool-design

Expert framework for designing agent-facing tools, optimizing tool descriptions, enforcing contract-based APIs, and implementing architectural reduction for reliable AI agent tool selection.

peon-ping-log

Track exercise reps like pushups and squats for the Peon Trainer. Log progress directly through your AI agent to trigger voice lines and stay motivated while coding.

test-delete-skill

Unified AI gateway for 100+ LLMs with OpenAI-compatible API, model fallbacks, load balancing, and enterprise-grade tools.

land

Automated PR lifecycle management: monitors conflicts, resolves CI failures, handles review feedback, and executes squash-merges for safe code integration.

gemini-audio

Implement Google Gemini API audio capabilities: process, transcribe, and summarize audio files, analyze environmental sounds, and generate natural speech with controllable TTS.

hegelion

Dialectical reasoning and adversarial coding agent for MCP-enabled editors, forcing LLMs to resolve internal contradictions for higher quality outputs.