AI & ML

constructive-io Bundle

Access control — roles, capabilities, profiles, grants, membership access, and entity-scoped authorization. Use when asked to 'assign capabilities', 'create roles', 'set up profiles', 'grant access', 'capability defaults', 'admin vs owner', 'membership capabilities', 'effective capabilities', 'revoke access', 'role hierarchy', 'custom roles', 'profile bundles', 'default profile', 'entity-scoped capabilities', 'org capabilities', 'app capabilities', 'capability resolution', 'trust levels', 'do owners get everything', or when working with the access control model in blueprints or the ORM.

blackwell-systems-agent-lsp Skill

First-session project onboarding. Explores the project structure, detects build system, test runner, entry points, and key architecture patterns. Produces a structured project profile the agent can reference throughout the session.

yanghu819 Skill

4090 一键跑 delta-embedding benchmark：本机缓存 HF -> rsync 上云 -> uv venv -> 产出 metrics.json（Template vs Content）。

coollabsio Skill

Use this skill for Laravel MCP development only. Trigger when creating or editing MCP tools, resources, prompts, or servers in Laravel projects. Covers: artisan make:mcp-* generators, mcp:inspector, routes/ai.php, Tool/Resource/Prompt classes, schema validation, shouldRegister(), OAuth setup, URI templates, read-only attributes, and MCP debugging. Do not use for non-Laravel MCP projects or generic AI features without MCP.

eternnoir Skill

Generate text embeddings using Google Gemini API for RAG, semantic similarity, classification, and clustering tasks. Invoke when user wants to embed text, create embeddings, or convert text to vectors with Gemini.

eternnoir Skill

Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.

eternnoir Skill

Self-iterating autonomous loop where the main agent acts as a pure orchestrator, spawning isolated Implementor and Critic subagents in rounds until the artifact converges, an iteration cap is reached, or human escalation is needed. Domain-agnostic — works for code, documents, plans, designs, copy, specs, proposals, or anything with an iterable artifact and explicit acceptance criteria. Also addressable by the short alias "bvl".

manykarim Bundle

Guide AI agents in creating native desktop UI tests with PlatynUI (new_core) using the PlatynUI.BareMetal library — cross-platform desktop automation over Windows UIA and Linux AT-SPI2 with an XPath query model. Use when asked to automate desktop applications (windows, buttons, menus, text fields), locate native UI elements by XPath, or write Robot Framework tests with PlatynUI.

tan-yong-sheng-ai-vision-mcp Skill

Analyze images and videos with AI vision models. Detect objects with bounding boxes, compare multiple images, audit design compliance, and analyze video content using Google Gemini or Vertex AI. Supports CLI and MCP modes.

tan-yong-sheng-ai-vision-mcp Skill

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 interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.

manykarim Bundle

Guide AI agents in creating native desktop UI tests with PlatynUI (new_core) using the PlatynUI.BareMetal library — cross-platform desktop automation over Windows UIA and Linux AT-SPI2 with an XPath query model. Use when asked to automate desktop applications (windows, buttons, menus, text fields), locate native UI elements by XPath, or write Robot Framework tests with PlatynUI.

sounder25 Bundle

Provides a Model Context Protocol (MCP) server interface to the skills library, allowing any MCP-compliant agent (e.g. Claude Desktop) to invoke Antigravity skills as native tools.

sounder25 Bundle

Wraps command execution in a self-healing loop. Uses heuristics to auto-fix common errors (missing deps, locked files) or escalates to the Agent with structured error context.

sounder25 Bundle

Detects when the agent is stuck in a reasoning loop or unproductive state by analyzing tool usage and sentiment patterns.

sounder25 Bundle

Updates standardized intent files (NEXT.md, STATE.json) to maintain context across agent sessions.

latitude-dev Skill

Create clear, actionable GitHub issues for bugs, features, and improvements. Issues are primarily consumed by LLMs, so optimize for agent readability and actionability.

sounder25 Bundle

Prepares isolated sub-workspaces for parallel agent execution. Copies context and generates specific mission instructions for "Worker" agents.

elie222 Skill

Guidelines for implementing LLM (Language Model) functionality in the application

elie222 Bundle

Guidelines for testing the application with Vitest, including unit tests, integration tests (emulator), AI tests, and eval suites for LLM features

elie222 Skill

End-to-end feature testing — browser QA, API verification, eval tests, or any combination. Covers browser interactions (via agent-browser CLI), Google Workspace operations (gws CLI), API calls, and LLM eval tests. Can also persist checks as reusable automated tests.

elie222 Bundle

Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to a site", "automate browser actions", or any task requiring programmatic web interaction.

latitude-dev Skill

Adding or changing API operations in `@repo/operations`. One source of truth (`defineOperation` + a Zod schema) becomes an HTTP endpoint, an OpenAPI operation, an MCP tool, TS + Python SDK methods, a `latitude` CLI command, and an in-process agent tool — descriptions and contracts must be written with all of these readers in mind.

elie222 Bundle

Review prompt and tool-description changes in a branch or PR. Use when asked to audit system prompts, prompt builders, tool descriptions, prompt inputs, or nearby evals/tests; summarize what changed; identify guidance that feels eval-shaped or too heavy for product use; and find repeated instructions between the main prompt and tool descriptions.

fatihkan Skill

Entry point and router for the Badi skill collection. Lists every bundled Badi skill grouped by category. Triggers when an agent needs to discover Badi capabilities or pick the right skill for a task.

fatihkan Bundle

Badi's behavioral discipline distilled into a portable skill — surgical changes, simplicity first, think before coding, goal-driven execution, yak-shave detection, TaskBoard discipline, knowledge-base source requirement, destructive-action gate. Extracted from Badi's 21 subagents, 12 hooks, and project CLAUDE.md. Triggers on: coding, refactoring, code review, writing new features, editing existing code, LLM coding pitfalls, overcomplication, bloated abstractions, missing tests, unclear requirements, scope creep, yak-shaving, dead code, premature optimization, technical debt, surface assumptions, verifiable goals.

creator-hian Bundle

Use when a developer needs to choose between two or more approaches during implementation. Triggers on "A vs B?", "should I use X or Y?", "which approach is better?", "Redis or Memcached?", "REST or gRPC?", "single table or split?", or any binary/multi-option technical decision. Use this skill whenever the user is weighing alternatives mid-implementation — even if they don't say "decide", if they're comparing approaches, this skill applies.

diegopherlt Skill

Harden a spec-driven specification until its design is closed — no missing, ambiguous, or unverifiable design decisions remain. Runs an autonomous loop-until-dry under the /goal command: each round spawns a FRESH closed-design-enforcer sub-agent that detects net-new design gaps, which this skill deduplicates and logs into the spec's managed Design Gaps section, until a round finds nothing new. Then it runs a single human arbitration gate. Use this whenever the user wants to "close the design", "harden the spec", "find design gaps", "make the spec ready for cold execution", or before implementing any spec via implement-spec. Invoke proactively once a spec draft exists and before implementation begins.

diegopherlt Skill

Implement a closed spec file end-to-end, dispatching to the execution strategy that fits the work. Designed to run alongside the /goal command, which drives it non-stop until the spec is implemented, quality-reviewed, and verified to actually work. Acts as a pure dispatcher: it separates strategy selection from execution, choosing write-directly, agent-waves, or workflow based on scale and orchestration closure. Use after a spec's design is closed (via close-design) and the user wants it built. Trigger on "implement the spec", "build this spec", "execute the spec", or a /goal aimed at a spec file.

diegopherlt Bundle

This skill should be used when the user says "refactor this", "fix this smell", "apply Extract Method", "apply [technique name]", "clean up this code", or passes a finding code from smell-scan (e.g. B1, D3, C2). Also trigger when the user wants to apply a specific refactoring.guru technique to a specific location, or to apply everything (or everything in a domain) flagged by a prior smell-scan. Carries each finding's technique, file, location, and smell context into a safe test → refactor → test → commit cycle, delegating edits to the refactoring-applier agent.

studiogadget Bundle

GitHub Copilot用のスキル（Agent Skills）を新規作成・改修する際に使用。「新しいスキルを作りたい」「既存スキルを改善したい」という依頼時に起動。要件ヒアリング→段階的開示設計→SKILL.md作成→リソース設計→検証の流れで進める。GitHub Copilotのコンテキスト制限（トークン上限）を考慮した簡潔な設計を重視。

creator-hian Bundle

Use when you have an existing plan, design, or implemented code that needs adversarial review to find weaknesses, missing edge cases, and unnecessary complexity

creator-hian Bundle

Use when creating implementation plans for complex multi-file features or architectural changes that benefit from multiple perspectives before implementation. Use this skill whenever the user asks to plan, design, or architect a feature that touches 3+ files, involves trade-offs between approaches, or would benefit from thinking through risks and requirements before coding. Even if the user says "plan this" without mentioning perspectives, this skill applies.

getty104 Skill

Create a new sub-agent definition file (agents/*.md) for the Claude Code plugin. Use this skill when the user wants to add a new specialized agent, define a new agent type, or create an agent that handles a specific domain of tasks. Trigger whenever the user mentions creating, adding, or defining a new agent or sub-agent.

diegopherlt Bundle

Guidelines for prompting and building context for Claude 5 generation models (Sonnet 5, Opus 5, Fable 5), based on Anthropic's official guidance from July 2026. Use this skill whenever the user asks how to prompt Claude 5 models, wants to write or review a system prompt, CLAUDE.md, skill, or agent harness targeting these models, mentions "unhobbling", "unknowns", "blind spot pass", "context engineering", or asks why an agentic task came back wrong despite detailed instructions. Also use it when auditing existing prompts or rules for over-constraint, or when migrating prompts written for Claude 4-era models. When invoked with file path arguments, audit those files against the guide (prefer delegating the audit to a read-only sub-agent and returning only findings).

diegopherlt Bundle

Designs, writes, or audits dynamic workflows for Claude Code — the JavaScript orchestration scripts the Workflow tool runs to fan many subagents across one large job (triggered by 'ultracode' or by asking to "use a workflow"). Use this skill proactively whenever the user says "create a workflow", "write a dynamic workflow", "orchestrate subagents", "fan out agents", "parallelize this with agents", "codebase-wide audit", "massive bug hunt", "ultracode workflow", "port this from X to Y", "migrate the whole repo", or describes work a single pass can't hold (repo-wide audits, large migrations, language ports, verified multi-source research). Also use it to audit or review an existing workflow script — "audit this workflow", "review my orchestration script", "is this workflow correct", "improve this workflow" — checking data-flow shape, per-role thinking load, and runtime safety. For authoring a single isolated sub-agent that reports back once, this is not the skill. Reach for this even when the user only hints at

diegopherlt Bundle

User-invoked shortcut to resume a multi-session run from its binnacle: locates .claude/binnacle.md at the current worktree root, delegates to the binnacle skill's resume procedure in full, optionally warms context with read-only explorers when the latest entry points at code it does not explain, and continues the work. Invoked manually as /resume-from-binnacle at the start of a cold session, to avoid retyping the resume prompt every time.