Explore AI Agent Skills & Claude Prompts
Discover open-source agent skills for Claude Code, Codex, ChatGPT, and any tool that uses SKILL.md.
Enter through keywords, occupations, creators, and GitHub sources to see what kinds of skills are emerging across domains.
Use the same catalog through the API
Connect 381,784 public skills to your own search, analytics, or agent workflow with the REST API.
Querying local SQLite index...
stag-strategy
by Senpi-aiSTAG v1.0.0 — Parabolic-Run Hunter. The entry-side pair for the parabolic_runner DSL preset. Enters LONG only when ALL FIVE gates pass: structural trend (close > 200-bar 4h SMA + 7d high within 48h), 7d move ≥ 25% (the parabolic threshold), 1.5x volume surge, acceleration (4d ≥ 7d/2), AND Smart Money LONG ≥ 60%. Built for HYPE-class runs (+60% over 16 days) where standard DSL trails would chop on 5-8% intraday gyrations. Operator-driven: most ticks return empty by design.
lynx-strategy
by Senpi-aiLYNX v1.0.0 — Adaptive MIN_SCORE Self-Tuner. The Vulture v4.1 story productized into a first-class agent pattern. Runs a simple BTC/ETH/SOL/HYPE momentum scorer; every 6h pulls its own closed-trade history via audit_query, buckets trades by score, and auto-raises MIN_SCORE if any bucket at-or-above the floor is bleeding. The same operation Vulture's agent ran by hand, but on a scheduled cron. NEW archetype #15: Self-tuning / adaptive-threshold agent.
thesis-fund-strategy
by Senpi-aiTHESIS FUND v1.0 — one configurable engine, many one-tap macro bets. You bring the market VIEW; the fund expresses it with discipline. The THESIS env var selects a preset from thesis-presets.json (e.g. risk_off = bet against the Trump economy, recovery = US risk-on, war_escalation / war_recovery = Iran/US/Israel, hype_vs_market, gold_over_btc / btc_over_gold). Each preset is a long basket + a short basket that together express the view, held in ONE wallet — but it is NOT a blind hold: each name is only pressed when the market is CONFIRMING the thesis direction, and the DSL + drawdown gate de-risk when it isn't. NOT a copy-trader; the runtime owns the LLM gate (pass-through), DSL exits, and all risk.guard_rails.
dog-strategy
by Senpi-aiDOG v3.0.0 — Contrarian Pup (SM Exhaustion Fader), senpi_runtime_helpers migration. Plumbing-only flip from openclaw-CLI subprocess + mcporter subprocess + Python state to in-process SenpiClient (direct HTTPS for MCP, direct HTTP POST to runtime /signals, long-lived producer_daemon). Thesis preserved verbatim from v2.5: multi-asset (BTC/ETH/SOL/HYPE) contrarian fader, 3.0% exhaustion gate, regime hard-gate, 15m freshness gate, MIN_SCORE 8, conservative leverage (7x base, 10x at score 12+), wide DSL for reversal development.
camel-strategy
by Senpi-aiCAMEL v1.0 — Carry Hedge Fund. Two single-direction books on two wallets, one producer, harvesting funding carry. The HARVEST book shorts the most-positive-funding names (longs pay shorts → short collects); the PAYOUT book longs the most-negative-funding names (shorts pay longs → paid to hold). Both are gated to EXHAUSTING crowds (never a fresh trend against the carry), with trend/RSI as confirmation + risk control. The edge is carry — a structural, recurring funding inefficiency — not market direction. NOT a copy-trader: each book scores its own universe and pushes signals; the runtime owns the LLM gate (pass-through), DSL exits, and all risk.guard_rails. CAMEL_LEG env selects the book.
falcon-strategy
by Senpi-aiFALCON v1.0.0 — Conversion-Event Momentum (XYZ Pre-IPO → equity). A Pre-IPO Perpetual (IPOP) carries a structural funding signature (|funding| <= ~1e-7, max_leverage <= 5) and is price-throttled by trade.xyz Discovery Bounds. When the company IPOs the product CONVERTS to a standard equity perp — funding jumps ~100x, the leverage cap lifts, the throttle is removed — opening free price discovery. Falcon detects the IPOP→STANDARD flip and trades the post-conversion momentum. Distinct from Lemur, which trades the IPOP basket while it is still an IPOP. Wide let-winners-run DSL, 7d hard_timeout.
meerkat-strategy
by Senpi-aiMEERKAT v1.0.0 — Momentum-Event Sniper. Watches the Senpi momentum-event feed (leaderboard_get_momentum_events — the 4h rolling-window momentum / rank-jump events) and snipes the freshest, highest-tier events the moment they fire, entering in the momentum direction before the move is broadly known. Tier (event magnitude) + freshness gate the entry; SM alignment + rising volume are confirmation bonuses. Distinct from the Striker / rank-jump agents (Jaguar/Orca/Roach) which score a leaderboard universe. Let-winners-run DSL with a short hard_timeout.
senpi-trading-runtime
by Senpi-aiThe Senpi Trading Runtime OpenClaw plugin runs automated trading strategies on Hyperliquid end-to-end: external producers push signals over POST /signals, rule-based or LLM-gated actions decide whether to open positions, declarative risk guard rails enforce daily caps and drawdown halts, FEE_OPTIMIZED_LIMIT orders execute maker-first with optional taker fallback, the position_tracker scanner detects on-chain position changes (including positions opened manually or by other tools), and the DSL exit engine applies two-phase trailing stop-loss protection. A bundled stdlib-only Python Producer SDK (senpi_runtime_helpers — SenpiClient, producer_daemon, scanner_lock, tick_cache, parallel) is the canonical way to author push producers; a senpi-helpers operator CLI manages running daemons. Use when a user needs to create/install/list/delete runtime YAMLs, configure DSL phases/tiers/time-cuts, set up an external_scanner with an LLM decision gate, declare risk guard_rails, inspect DSL-tracked positions, check runtime +
otter-strategy
by Senpi-aiOTTER v2.0.0 — Open Interest Velocity Hunter (senpi_runtime_helpers migration). Plumbing-only port from v1.0. NO thesis change. NO scoring change. Producer ports onto `senpi_runtime_helpers` (in-process SenpiClient + direct HTTP POST to runtime /signals + producer_daemon long-lived loop). Trades the rate of change of OI on Hyperliquid perps — when 1h OI delta is >= 5% AND price moves in the same direction by >= 0.5%, that's fresh leveraged positioning with directional conviction (TOP-LEFT or TOP-RIGHT quadrant of the OI/price matrix). Otter follows the flow for 1-3 hours then exits. Conviction-scaled leverage (5/7/10 by score) per the fleet-winning pattern.
condor-strategy
by Senpi-aiCONDOR v4.0.1 — One Amazing Trade per Day. Top 50 HL assets, pure trend continuation, apex confluence only. 3TF alignment hard gate + MACRO_TREND_GATE + SM consensus >=70%, MIN_SCORE 12, score-scaled sizing (50%/70%/80%), 10x leverage cap, 6-tier DSL ladder from Kodiak SOL empirical wins. v4.0.1 ships a race-window dedup cache that eliminates the ENGINE_FAILURE retry noise on already-held assets, plus a doubled DSL exit interval to throttle REDUCE_ONLY spam when runtime position-state lags HL.
turbine-strategy
by Senpi-aiTURBINE v3.3 — Volume rotation on TWO wallets, ONE producer daemon. Both wallets receive the same volume-rotation alpha (same scoring, same asset universe, same funding-fade direction selection). The wallet boundary just selects DSL behavior: VOLUME wallet's runtime has hard_timeout 10min and no Phase 2 (pure rotation cadence); RUNNERS wallet's runtime has hard_timeout 240min and Phase 2 ratchet enabled (let winners run). Most positions exit at small loss/win on either wallet; ~5% land on a real directional move and ratchet to apex on the runners wallet — that asymmetry is the alpha v3.0/v3.1 was leaving on the table by force-cutting at 10 min. v3.3 doubles margin/slot to scale from v3.2's verified $2M/day to a $3-4M/day target while preserving the <$150/$1M cost efficiency on the same 7-asset tight-spread universe.
kodiak-strategy
by Senpi-aiKODIAK v7.0.0 — SOL alpha hunter, senpi_runtime_helpers migration. Plumbing-only flip from openclaw-CLI subprocess + mcporter subprocess to in-process SenpiClient (direct HTTPS for MCP, direct HTTP POST to runtime /signals, long-lived producer_daemon). Thesis preserved verbatim from v6.0.1: SOL alpha hunter, single-asset focus, v5.1 base-tech-score floor with multi-factor scoring (SM consensus, trend structure, momentum, funding, OI, BTC correlation, RSI), conviction-tiered leverage (5x default, 6x conviction at score 11+, 7x apex at score 13+), 240-min asset cooldown.
Browse Agent Skills by Occupation
23 major groups · 867 SOC occupations
Browse by Category
Explore agent skills organized by their primary use case
Explore the agent skills ecosystem by occupation and creator
SkillMD is not just a keyword search box. It is an open map that organizes public skills by occupation, creator, and repository, helping you see which workflows, judgment criteria, and domain habits people are writing for AI agents.
Then follow creators and GitHub repositories back to the source: compare the skills a team maintains, whether the repo is active, and how the README frames the work before you open, install, or reuse anything.
Use it three ways: learn an unfamiliar field by occupation, study how creators organize skills, then use source context to decide what is worth opening or reusing.
01 Map a field
Browse 23 occupation groups and 867 SOC roles to learn what skills exist in adjacent domains and how they break down real work.
02 Follow creators
Use creator and repository pages to inspect maintained skill collections, recent updates, and source context before trusting a result.
03 Search with sources
Search 1.7M+ collected skills, then use occupation tags, creators, and GitHub source context to decide what is worth opening.
Start with the occupation map, then follow creators and repositories back to real code. SkillMD helps explain why a skill is worth opening, not only what it is named.
Standardizing Agent Capabilities with SKILL.md and Model Context Protocol (MCP)
In the rapidly evolving landscape of artificial intelligence, LLM agents (Large Language Model agents) have transitioned from simple text predictors to autonomous problem solvers. To orchestrate complex, multi-step agentic workflows, developers require a standardized format to specify agent capabilities, prompt instructions, system rules, and database bindings. This is where SKILL.md and the Model Context Protocol (MCP) have emerged as standard developer paradigms. SkillMD serves as the central directory for indexing, exploring, and sharing these critical agent configurations.
Our open-source registry currently tracks over 1.7 million collected SKILL.md configurations and system prompts. By compiling agent configurations from active developers on GitHub, we bridge the gap between prompt engineering research and production execution. Whether you are building agents with Anthropic's Claude Code, OpenAI's GPT-4, Google's Gemini, or local models using Ollama and LlamaIndex, standardized skill definitions ensure your agents behave predictably across different runtime environments.
What is the Model Context Protocol (MCP)?
The Model Context Protocol (MCP) is an open-source standard designed to connect LLMs to data sources, developer tools, and external environments. MCP establishes a bidirectional communication channel between client applications (like Cursor, Claude Desktop, or custom agent systems) and servers hosting data or capabilities. Standardizing instructions via SKILL.md enables LLMs to query databases, read local files, execute terminal commands, and integrate third-party APIs. SkillMD allows you to find ready-to-run MCP servers and prompt instructions for various occupations and technical tasks.
The Structure of a Professional SKILL.md File
A valid SKILL.md configuration is designed to be easily read by humans and parsed by LLMs. It contains precise system instructions, trigger conditions, required parameters, and execution examples. Below is the typical architectural blueprint of a professional agent skill:
- Metadata & Core Scope: Declares the name of the skill, author details, target models, and a description of the capability.
- Triggers & Intent Detection: Details semantic triggers that help the agent decide when to invoke this skill.
- System Prompts: Explicit system-level instructions that direct the agent's behavior, personality, safety guardrails, and formatting preferences.
- Capabilities & Tools: Lists the files, databases, or APIs the agent must access to complete the tasks.
- Few-Shot Examples: Demonstrates real inputs and outputs, helping the model generalize behavior through in-context learning.
Optimizing Agent Workflows for Modern LLMs
Writing effective agent skills requires deep knowledge of prompt engineering. With the release of advanced reasoning models like Claude 3.5 Sonnet, ChatGPT o1, and DeepSeek-V3, prompt templates must focus on structured thinking. Developers are encouraged to use XML tags (e.g., <thought>, <context>, and <rules>) to isolate execution boundaries. Standardized prompts prevent agents from suffering from context drift, ensuring that long-running tasks remain aligned with the initial system parameters.
Exploring by SOC Occupations and Creator Profiles
What makes SkillMD unique is its taxonomy. Instead of simple text search, we parse and organize files according to the Standard Occupational Classification (SOC) system. This means you can discover skills written for Computer and Mathematical roles, Business and Financial operations, Legal, Design, and and Educational Instruction fields. By tracking creator profiles, developers can study how different teams organize their custom instructions, compare version updates, and fork public configs for specialized enterprise use cases.
SkillMD operates as a high-performance index running on a fast Go backend and a highly responsive Astro SSR frontend. All search queries execute in milliseconds, featuring smart debouncing to prevent multiple API requests while keeping user data secure. Join our community of developers to standardize your AI agent instructions and optimize your LLM prompting workflows today.
Frequently Asked Questions
A practical guide to agent skills: what they are, how to inspect them, and how SkillMD helps you explore the ecosystem.