Design a reusable agent loop as a graph before running it. Take a TASK and COMPLETION CRITERIA, draw the node graph (input, plan, act, verify, retry or escalate, done), specify context, tools, expected output, and evidence per node, label every transition trigger, set max retries, budget ceilings, and human approval…
Build, debug, and optimize Claude API / Anthropic SDK apps. Apps built with this skill should include prompt caching. TRIGGER when: code imports anthropic/@anthropic-ai/sdk; user asks to use the Claude API, Anthropic SDKs, or Managed Agents (/v1/agents, /v1/sessions, /v1/environments). DO NOT TRIGGER when: code…
Build and deploy autonomous agents — scripts that run on a schedule, call LLMs, scrape the web, store data, and act without user intervention. Use when user says "build me an agent", "create a bot", "automate X on a schedule", "monitor this page", "summarize X every day", "alert me when Y changes", or wants to deploy…
Use this skill when working on b3sty RedM/FiveM resources or related Lua code. The defaults below are the high-frequency rules applied on every task; open the reference files only when the task touches that area.
Maps Figma node hierarchy (FRAME/GROUP/INSTANCE) to UIKit UIView containers, preserving parent-child relationships, addSubview order, and coordinate system. Decides which nodes become UIView containers vs leaf views, and how to read relative/absolute coordinates from the data package. Use when generating UIKit code…
A tool for generating the starting structure of control firmware for MSPM0 or STM32 hardware, with optional K230 vision communication. Firmware is the software that runs directly on a device or microcontroller.
A thinking framework that breaks work into five stages: define the problem, explore options, formulate a solution, act, and assess the result. It can be applied fully, partially, or minimally depending on the task.
Use only in coordinator sessions to orchestrate planning or execution of work, including execute, continue, or resume-plan requests. Covers role definition, specialist discovery, dispatch, review, learning, and synthesis. Never use in Session type: DELEGATED executor sessions; those agents perform their handed scope…
DeepVista CLI — knowledge base, notes, chat, and skills from the terminal. One skill for the full deepvista CLI. Load when the user wants to: take a note, jot this down, save a fact, list / edit notes; create, search, pin, archive, edit, or grep knowledge-base cards (person, topic, file, note, todo…) via deepvista…
Use when organizing a large local skill library into category router skills, refreshing a compact skill inventory, or deciding which leaf skill bodies need to be read before agent classification.
Use when the user asks for a deep review, full review, comprehensive review, production readiness assessment, full audit, multi-domain audit, "security and reliability and code review", or "review everything". Also use when the user explicitly requests performance review alongside the comprehensive request (e.g.…
Use when working with RisingWave — streaming SQL database, materialized views, sources, sinks, CDC from PostgreSQL/MySQL, Kafka ingestion, time windows (TUMBLE/HOP/SESSION), watermarks, EMIT ON WINDOW CLOSE, port 4566, risingwave-mcp, event streaming pipeline design, real-time analytics, stream processing.
A library of UI design direction files that provide style guardrails for building tasteful, consistent interfaces. Use when building any UI — select a style or let the AI recommend one based on the project context.
Use when the target is rare (fraud, churn, disease, anomalies). Covers correct metrics, resampling (SMOTE/undersampling), class weights, threshold tuning, and avoiding the accuracy trap and resampling leakage.
Create a new source-grounded decision brief in the source format by default (Feishu/Lark, Word/docx, Markdown, or self-contained HTML) with a reader-fit Pyramid opening, one editable or auditable visual covering at least 80% of value-weighted non-appendix claims, and one key-question table. Use when the user names…
Use the GitHub CLI to fetch content from GitHub URLs instead of web fetching. Activate when the user shares a GitHub issue, PR, repo, or discussion link.
Create short AI-901 practice labs (15-25 minutes) that are executable and self-validating. Every lab includes prerequisites, exact tasks, validation steps, expected outputs, and cleanup. Scope covers Microsoft Foundry portal builds, Foundry Python SDK client apps with keyless authentication, Azure AI Services via…
★not rated 9 2mo agoA97 tokens
originalMIT
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: