All nine kinds in one list: what an agent loads, what it runs, what it
reads and what bundles the rest. 436,218 of
them, each with its token cost per session, a security scan with the
line that triggered it, and whether it is the original or one of the
copies.
An AI-powered senior engineering lead that conducts comprehensive code reviews. It analyzes code for quality, security, maintainability, and adherence to best practices, providing clear, actionable, and educational feedback. Use immediately after writing or modifying code.
Your habits, imprinted on AI, plus judgment that adapts to you: when to act, when to check first, when the project rules win. A portable working profile across every agent. Powered by I-Lang v5.0.
★not rated 104 1mo agoA
tokens not measured
originalMIT
MCP server "ue-llm-toolkit" as configured in ColtonWilley/ue-llm-toolkit. Runs Resources/mcp-bridge/index.js with node. Needs 1 environment variable to run.
★not rated 104▲
+1 7mo agoA
tokens not measured
originalMIT
A stock-analysis assistant for Chinese, Hong Kong, and United States stocks that examines price charts, technical indicators, money flows, overseas-investor holdings, and dividends.
A lightweight RSS subscription manager for tracking updates from services such as YouTube, Vimeo, Behance, Twitter/X, and several Chinese platforms. RSS is a format that lets one tool collect new posts from many sites.
Orchestrates continuous kernel optimization by chaining profiling, bottleneck diagnosis, tier-based optimization, verification, and benchmarking into an iterative loop. This is the "main program" that drives the other skills. Use when you want to iteratively optimize a kernel until it meets a performance target…
Starter skill for analyzing Lu.ma events, attendees, invites, and engagement signals from organizer-provided API access, exports, or manual event inputs. Use this when a template or workflow needs a repeatable path to turn Lu.ma event data into an organizer-ready analysis brief.
A set of rules for AI-assisted coding, including Chinese-language communication and comments, commit-message formatting, small changes, and human review of sensitive code.
Create, review, debug, and recover LS-DYNA keyword decks and simulation workflows. Use when working with LS-DYNA .k or .key input files, include trees, explicit crash or impact models, implicit static or quasi-static solves, contact definitions, section and material selection, timestep or mass-scaling control…
Audit and improve existing CLAUDE.md files across a repository. Use for a deliberate, repo-wide quality review—not to record learnings from the current session.
A visual-analysis skill that studies a reference image and turns its visible structure, composition, colors, materials, and style into a reusable Chinese image prompt.
Design intelligence engine for AI coding. v3 ships THE BRAIN: a deterministic 7-axis synthesizer (warmth, contrast, density, geometry, formality, motion, typepersonality) that compiles a novel design language per brief: no LLM calls, fully offline, closed-loop learning via a local decisions ledger. 1,243 structured en.
Direct a website the way a film is made: cast a director and a film, choose the medium the film calls for (photoreal 3D, hand-drawn, ink wash, print, pixel, neon, type, photography…), cast a lead, ask for the assets, write a shot list for every section, then shoot it with a scroll-driven camera, designed cuts, a…
Use durable, explainable, self-improving memory across a coding session via the velesdb-memory MCP server. Trigger whenever the velesdb-memory MCP tools (remember/recall/recallfused/relate/why/feedback/forget/rememberextracted/extractionstatus/entity/memorystatus) are available and the work would benefit from…
A toolkit for writing Chinese online novels from early ideas through revision and publication checks. It covers story planning, drafting, character and world design, continuity checks, pacing review, and text editing.
★not rated 104▲
+6 7d agoA
tokens not measured
originalMIT
A tool that delegates coding, codebase exploration, implementation, review, and verification work to a persistent Codex CLI agent. It is intended for requests that explicitly ask Codex to act or continue an existing Codex task.
MCP server "experience-hub", hosted remotely at localhost, as configured in zhangjunmengyang/quant-research-platform.
★not rated 104 7mo agoA
tokens not measured
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: