LLM skills

9,876 tagged LLM, measured the same way as everything else here.

Browse within: agent 502chatgpt 227agents 211cli 169ml 133bun 129javascript 124agentic-ai 122openai 118agent-framework 116deep-learning 107foundation-models 98agentskills 90cortex 90

codex

553

StarryCod/cogitum

Skill Codex

Delegate coding to OpenAI Codex CLI (features, PRs).

not rated 11 3mo ago A 17 tokens original MIT

espai

554

pouriya/espai

Skill Claude CodeCodex

Drive the user's real browser — their tabs, their logged-in sessions, their live DOM — through a plain HTTP server on 127.0.0.1, using nothing but curl. Works in any Chromium-engine browser: Chrome, Edge, Brave, Chromium, Opera, Vivaldi. Use whenever a task needs a real browser rather than a fetch: reading a page…

not rated 11 1mo ago A 115 tokens original MIT

frontend-design

555

victorzhong0110/skill-evolution

Skill Claude CodeCodex

Produce a distinctive, subject-grounded UI plan and implementation notes. Use when the user asks for a new page, landing, dashboard, or restyle — including requests that never say "design system" but clearly need visual direction.

not rated 11 +1 6d ago A 49 tokens original MIT

codebase-second-brain

556

Mallikarjun-Roddannavar/playwright-agentic-automation

Skill Codex needs its repo

Retrieve, verify, and refresh the persistent Open Knowledge Format v0.2 codebase memory for playwright-agentic-automation. Use when an agent needs architecture discovery, source ownership, page/API/fixture relationships, static AST graph queries, knowledge freshness checks, Mermaid/Obsidian navigation, or durable…

not rated 11 6d ago A SkillSpector: pass 72 tokens original MIT

xquik-social-data

557

librefang/librefang-registry

Skill Claude CodeCodex

Use Xquik for X and Twitter social data workflows through its public API, SDKs, MCP server, webhooks, and installable agent skill.

not rated 11 18d ago A SkillSpector: pass 37 tokens original MIT

vandalizer-api

558

ui-insight/vandalizer

Skill Claude Code

Talk to the Vandalizer Management API — /api/mgmt/v1. Use when the user says "/vandalizer-api", asks to query the management API, pastes a vklive… key, asks to check Vandalizer dev/prod stats, validation runs, workflows, audit log, or wants to run a workflow/validation/extraction via the management surface.

not rated 11 yesterday A 84 tokens GPL-3.0

speckit-archive

559

opsmill/infrahub-mcp

Skill Claude Code

Archive a feature specification into main project memory after merge, resolving gaps and conflicts.

not rated 10 yesterday A SkillSpector: pass 21 tokens original Apache-2.0

cost-audit

560

berkcangumusisik/claude-code-practices

Skill Claude Code

Analyzes the current session's token usage, pinpoints where tokens were wasted, and produces concrete optimization actions.

not rated 10 4mo ago A 25 tokens original MIT

pydantic-ai-agents

561

Fuenfgeld/pydantic-ai-skills

Skill Claude CodeCodex

Build and debug Pydantic AI v2 agents using best practices for dependencies, instructions, tools, capabilities, hooks, and structured output validation. Use when the user wants to: (1) Create a new Pydantic AI agent, (2) Debug or fix an existing agent, (3) Add features like tools, validators, capabilities, hooks, or…

not rated 10 2mo ago A 135 tokens original MIT

SummerRiversound/human-readable-reports

Skill Claude CodeCodex

Use when writing an analysis, report, findings, or status doc for humans to read and share — not a spec an agent executes. Fights AI verbosity/slop and makes deep content easy to read.

not rated 10 1mo ago A 46 tokens original MIT

PheroPath

563

zi-yue-1129/PheroPath

Skill Claude CodeCodex

A filesystem-based stigmergy communication protocol. Use this to leave invisible "pheromones" (signals) on files to communicate context, risks, or status to other agents or your future self without modifying file content.

not rated 10 7mo ago A 48 tokens original MIT

elon

564

Democraitus/elons-brain

Skill Claude CodeCodex

Talk to Elon Musk. Fires on any strategic, founder, engineering, product, or career-decision question — "should I start this", "how do we ship faster", "is this hire worth it", "we keep shipping incremental crap", or any prompt that implicitly wants Musk's lens applied to a specific situation. This is a self-contained…

not rated 10 3mo ago B 204 tokens

deciqAI/knowledge-skills

Skill Claude CodeCodex

Diagnoses the buyer/seller deadlock in any trade of information — the buyer cannot value what they cannot see, but once they see it they have it for free — and selects the disclosure mechanism (patent, NDA, staged disclosure, trusted intermediary + escrow, reputation, or proxy demonstration) that lets the buyer…

not rated 10 10d ago A SkillSpector: warn 255 tokens original MIT

build-html-artifacts

566

amol-patil/html-artifact-kit

Skill Codex

Build polished, self-contained HTML artifacts. Always use for requests to create, design, improve, or validate a dashboard, report, presentation or slides, narrative document, microsite, portal, console, operational tool, webpage, quick side artifact, or any standalone HTML deliverable, even when the user does not…

not rated 10 20d ago A 95 tokens original MIT

tidal-cli

567

lucaperret/tidal-cli

Skill Claude CodeCodex

Control Tidal music streaming from the terminal. Use when the user wants to search Tidal's catalog (artists, albums, tracks, videos, playlists), manage playlists (create, rename, delete, add/remove tracks), manage library/favorites, play music, explore artist/track info, find similar artists or tracks, get…

not rated 10 2mo ago A 130 tokens original MIT

data-analysis

568

yujiachen-y/agent-bundle

Skill Claude CodeCodex

Analyze data using Python with pandas, matplotlib, and numpy.

not rated 10 2mo ago A 15 tokens original MIT

pachca-bots

569

pachca/openapi

Skill Claude Code

A local MCP server for creating diagrams with the GenAI DrawIO Creator package. MCP is a way for an agent to call an external tool or service.

not rated 10 yesterday A SkillSpector: warn 192 tokens original MIT

hermespace

570

PabloTheThinker/hermespace

Skill Claude CodeCodex needs its repo

Full Hermespace pocket workbench for Hermes Agent (v0.18+): concept, dual decode, workbench I/O, fabric skills+MEMORY, FOA desk, Pulse, viewport app-shell, Tailscale serve, pocket boundary, autonomy grid, Desktop plugin, ops boot/doctor/smoke. Use for hermespace, pocket, workbench, FOA, pulse, viewport, access…

not rated 10 21d ago A SkillSpector: warn 100 tokens original MIT

summarizer

571

JiauZhang/chatchat

Skill Claude Code

A Chinese-language instruction for summarizing long text when the user asks for a summary or overview.

not rated 10 21d ago A 20 tokens GPL-2.0

sign-draft

572

sign-protocol/sign-tools

Skill Claude Code

Author a new SIGN document from a concept, notes, or requirements — interactively and spec-compliant. Use when an operator asks Claude Code to write, draft, or create a SIGN document (a std / pol / ref / proc / xwalk doc), produce a Layer 1 index entry, or hand-write SIGN structure. For converting an EXISTING source…

not rated 10 1mo ago A 86 tokens original Apache-2.0

explainer-site

573

sciclaw/sciclaw

Skill Claude CodeCodex

Create "How X Works" deep-dive explainer websites with the curious, technical-but-accessible style of howkeyboardswork.com. Use when asked to explain how something works, create an educational single-page guide, or build an interactive explainer site on any topic.

not rated 10 2mo ago A 59 tokens

research

574

nullskymc/Aurevoy

Skill Claude Code

A research skill for turning web searches into short reports or deeper, structured investigations. It can deliver the result in chat, Markdown, or HTML when a page-like layout is needed.

not rated 10 1mo ago A 62 tokens original MIT

short_novel_write

575

gyyxs88/short_novel_write

Skill Claude CodeCodex

A step-by-step skill for writing Chinese short stories and saving the finished manuscript as a local Markdown file. It supports planning, drafting, self-checking, and saving.

not rated 10 5mo ago A 64 tokens original Apache-2.0

unmachined

576

0xNyk/unmachined

Skill Claude CodeCodex

Finds and removes AI slop from prose, UX copy, and web interfaces. Use when writing, auditing, or revising articles, posts, READMEs, marketing copy, UI copy, layouts, or frontend code that must not feel AI-generated; when the user says unmachined, de-slop, AI slop, anti-slop, humanize, make this human, sounds like AI…

not rated 10 18d ago A 198 tokens original MIT

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: