AI & ML

Move issues and external PRs through a state machine of triage roles, categorise, verify, grill if needed, and write agent-ready briefs.

利用python，指定某个arxiv_id/url， 基于 LLM Agent 对这篇arxiv论文进行分类与深度阅读，直接print打印阅读笔记

和 AI 协作写小说的工作流系统。9 个 agent 协作完成从设定到归档的完整写作流程。入口检测 → 初始化/迁移 → 交 novel-agent 调度。适用场景：从零写新小说、导入已有小说。

Claims-based authorization for agents and operations. Grant, revoke, and verify permissions for secure multi-agent coordination. Use when: permission management, access control, secure operations, authorization checks. Skip when: open access, no security requirements, single-agent local work.

Agent skill for tester - invoke with $agent-tester

Agent skill for sandbox - invoke with $agent-sandbox

AI Manipulation Defense System with self-learning prompt injection detection and adaptive mitigation

V2 backward compatibility layer mapping underscore-based MCP tool names to V3 slash-based implementations with parameter translation

Agent skill for pr-manager - invoke with $agent-pr-manager

KV cache compression for LLM inference — 4.4x compression, 2x context capacity, near-lossless quality. ICLR 2026 paper implementation with vLLM integration.

Agent skill for swarm-issue - invoke with $agent-swarm-issue

Agent skill for github-modes - invoke with $agent-github-modes

Agent skill for safla-neural - invoke with $agent-safla-neural

Agent skill for code-analyzer - invoke with $agent-code-analyzer

Agent skill for load-balancer - invoke with $agent-load-balancer

Join a Google Meet call, transcribe live captions, optionally speak in realtime, and do the followup work afterwards. Use when the user asks the agent to sit in on a meeting, take notes, summarize, respond in-call, or action items from it.

Spotify: play, search, queue, manage playlists and devices.

MCP client: connect servers, register tools (stdio/HTTP).

HeartMuLa: Suno-like song generation from lyrics + tags.

Search/download GIFs from Tenor via curl + jq.

Karpathy's LLM Wiki: build/query interlinked markdown KB.

ASCII art: pyfiglet, cowsay, boxes, image-to-ascii.

GitHub auth setup: HTTPS tokens, SSH keys, gh CLI login.

DSPy: declarative LM programs, auto-optimize prompts, RAG.

Simple Preference Optimization for LLM alignment. Reference-free alternative to DPO with better performance (+6.4 points on AlpacaEval 2.0). No reference model needed, more efficient than DPO. Use for preference alignment when want simpler, faster training than DPO/PPO.

Read, search, create, and edit notes in the Obsidian vault.

Apple Reminders via remindctl: add, list, complete.

Benchmark LLM performance across tasks — latency, quality, cost comparison.

24/7 system health monitoring, auto-restart crashed services, and multi-channel alerting (Discord, Slack, email).

Design authentication and authorization strategy for a product or MVP. Use when the team needs sign-in methods, session model, user/workspace membership, roles, tenant boundaries, invitation flows, or permission enforcement plans.

Expert in CrewAI - the leading role-based multi-agent framework used by 60% of Fortune 500 companies.

Job search automation, auto apply, resume generation, application tracking, salary intelligence, and recruiter outreach using the JobGPT MCP server.

Turn a one-line objective into a step-by-step construction plan any coding agent can execute cold. Each step has a self-contained context brief — a fresh agent in a new session can pick up any step without reading prior steps.

Conducts multi-axis code review. Use before merging any change. Use when reviewing code written by yourself, another agent, or a human. Use when you need to assess code quality across multiple dimensions before it enters the main branch.

Design reproducible evaluations for AI agents with representative task sets, explicit rubrics, appropriate graders, baselines, regression gates, and failure analysis. Use when defining agent quality, comparing prompts or models, validating a release, measuring tool-use reliability, investigating regressions, or deciding whether an agent is ready for production.

Optimize machine learning model hyperparameters using grid search, random search, Bayesian optimization, and Hyperband to maximize model performance within a compute budget. Use when the user requests hyperparameter tuning or provides relevant inputs for this workflow.