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...
implementation-architect
by kxlahsimx09Implementation-architect for the next-generation Mobiz payment gateway (mb-next-payment-gateway). Materializes each ratified ADR as a cheap, runnable PoC + spec tests asserting ADR-promised claims + a drift report when execution falsifies a claim. Mines the current-system corpus (mobiz-payment-gateway + bank-bot, tagged #current) for realistic seeds — vault learnings, integration-tests, docs/flows — and binds them into spec-test docstrings and PoC fixtures. Lives between architect (upstream) and developer (downstream). Sibling — not replacement — to next-dev. Falsifier / prover, not designer or builder. Trigger this skill when the user says: "PoC for ADR-N", "validate this decision", "stand up a falsifying PoC", "spec-test the wallet ledger claim", "drift report on §ADR-4b", "is the ADR realistic?", "implementation-architect", "next-impl", "ทำ PoC", "พิสูจน์ ADR", "เทสต์การออกแบบ", or any request to convert a ratified ADR into a runnable falsifying experiment.
tester
by kxlahsimx09Software quality agent for the Mobiz Payment Gateway. Reads Go, Node.js, and MongoDB code directly, studies the integration-tests/ suite (including mock-bank/), and reports which tests are valid, stale, or mis-configured. Also proposes new test cases to close coverage gaps. Operates as a READ-ONLY auditor of production code — it may edit integration-tests/** and mock-bank/** but never the code under test. Supersedes the earlier `integration-test-writer` skill (see §11). Trigger this skill when the user says: "review tests", "validate tests", "check test validity", "ตรวจสอบเทส", "เทสยัง valid ไหม", "test is broken — why", "add test case", "เพิ่มเทสเคส", "mock-bank drift", "coverage gap", "tester", "QA", "regression", "stress test", "smoke test", or any request about the health of the payment-gateway test suite.
technical-writer
by kxlahsimx09Payment-gateway technical writer. Reads Go, Node.js, and MongoDB code directly, tracks commits, and writes/updates documentation so it always reflects the live system. Produces dual-audience docs — human onboarding plus agent-parseable structure — for both the CURRENT system (Go + Fiber + MongoDB + bank-bot) and the TARGET system being migrated to (code-only migration, fresh data). Owns `docs/`, `docs-site/`, `README.md` deltas, architecture diagrams, ADRs, runbooks, API references, and migration notes. Trigger this skill when the user says: "write docs", "update the readme", "document this", "เขียน doc", "อัปเดต doc", "doc ยัง sync อยู่ไหม", "technical writer", "tech writer", "ADR", "runbook", "release notes", "migration guide", "architecture diagram", "API reference", "explain the bank-bot flow", "what changed in commit X", or any request to reconcile documentation with actual code. Also triggers when a new commit or PR lands and docs need to catch up.
technical-writer
by kxlahsimx09Payment-gateway technical writer. Reads Go, Node.js, and MongoDB code directly, tracks commits, and writes/updates documentation so it always reflects the live system. Produces dual-audience docs — human onboarding plus agent-parseable structure — for both the CURRENT system (Go + Fiber + MongoDB + bank-bot) and the TARGET system being migrated to (code-only migration, fresh data). Owns `docs/`, `docs-site/`, `README.md` deltas, architecture diagrams, ADRs, runbooks, API references, and migration notes. Trigger this skill when the user says: "write docs", "update the readme", "document this", "เขียน doc", "อัปเดต doc", "doc ยัง sync อยู่ไหม", "technical writer", "tech writer", "ADR", "runbook", "release notes", "migration guide", "architecture diagram", "API reference", "explain the bank-bot flow", "what changed in commit X", or any request to reconcile documentation with actual code. Also triggers when a new commit or PR lands and docs need to catch up.
next-tester
by kxlahsimx09Quality / evidence agent for the next-generation Mobiz payment gateway (mb-next-payment-gateway). Forks the implementation-architect's `poc/integration/` harness (SPEED fast-clock + fixture-loader + per-story probes + SLO assertions) into the regression suite, builds one fixture + probe per story that quotes and asserts the story's AC clauses, maintains the test-index + coverage-gap log, and runs integration smoke (SPEED virtual-clock on real substrate) + perf. READ-ONLY on production code — never edits supabase/functions, migrations, or gateway code. Builds evidence; does NOT self-certify completeness (the investigator audits the evidence and issues the epic seal). Sibling to next-dev (builder, upstream) and next-investigator (skeptic, downstream). Trigger this skill when the user says: "build the probe for DEPOSIT-001", "run the smoke suite", "add a fixture", "coverage gap", "perf test the deposit flow", "next-tester", "เขียนเทส story", "รัน smoke", or any request about next-gen gateway test evidence.
next-ui
by kxlahsimx09UI Designer + Implementer for the mb-next backend-user web portal (kxlahsimx09/mb-next-admin-portal) — the console through which backend users (admin, client, sub-client, partner) operate on the next-generation Mobiz payment gateway (mb-next-gateway). Owns the LOOK and the UI IMPLEMENTATION of the portal — components, design tokens (DESIGN.md), screen and dashboard views — and drives them with the `impeccable` workflow (impeccable.style: a Claude-powered plan→build→review→refine loop that runs ON Claude Code). Does NOT touch backend EF/RPC/migrations, ADRs, tests, or per-story acceptance text. Trigger this skill when the user says: "design the UI", "build the component", "style the portal", "run impeccable", "design tokens", "critique this screen", "next-ui", "ออกแบบหน้าจอ", "ทำ component", or any request to design or implement the portal UI.
next-code-reviewer
by kxlahsimx09The REVIEW gate for the next-generation Mobiz payment gateway (mb-next-payment-gateway). Reviews EVERY next-dev PR on three dimensions — (1) ตรงตาม requirement: code implements the story AC fully + conforms to the ADR/substrate contract; (2) code clean: readable, DRY, ≤250 lines/file, no `any`, proper error handling, migrations not inline-SQL; (3) performance smells in the diff: N+1, missing index, unbounded loop, sync heavy work in hot path, EF cold-start, lock-ordering deadlock risk, EF 150s limit. Output is `gh pr review --approve | --request-changes` grouped by the three dimensions. This is the gate — next-dev cannot deliver while a `--request-changes` is open. Does NOT merge, does NOT edit code, does NOT run load tests. Distinct from next-investigator (which audits evidence-vs-claim). Trigger this skill when the user says: "review this PR", "review DEPOSIT-001", "is this PR ready", "next-code-reviewer", "รีวิว PR", "ตรวจโค้ด", or any request to gate a next-dev PR.
next-dev
by kxlahsimx09Builder for the next-generation Mobiz payment gateway (mb-next-payment-gateway). Implements each ratified story's acceptance criteria as production code on real substrate — Supabase Edge Functions (Deno/TS), Postgres migrations, and the gateway/worker layer — forking the implementation-architect's `[POC_PROMOTED]` PoC as the seed. Builds to story AC + ADR + substrate contract; opens one PR per story, story-id linked, and cannot deliver until next-code-reviewer approves (the REVIEW gate). Sibling — not replacement — to next-impl (upstream PoC) and next-tester (downstream evidence). Builder, not designer, not test-author. Trigger this skill when the user says: "implement DEPOSIT-001", "build this story", "promote the PoC to production", "write the edge function for", "next-dev", "เขียนโค้ด story", "ทำ feature ตาม AC", or any request to turn a ratified story into running code on the next-gen gateway.
next-investigator
by kxlahsimx09The Skeptic / Falsifier for the next-generation Mobiz payment gateway (mb-next-payment-gateway). Believes EVIDENCE ONLY, never claims — collides every claim against real evidence (evidence/integration-run-*.json + git-sha, logs, PR diff, #current vault, production data when live). Owns the VERIFY sub-gates V1 (audit each probe quotes + asserts its AC clause) and V5 (epic-close completeness audit — AC coverage vs INDEX + sample-probe rigor, the audit#141 pattern) and issues the "epic seal" that next-pm requires to mark an epic done. Can reopen stories. Runs its OWN independent full regression on its OWN seal env — does not trust the tester's env. Distinct from next-code-reviewer: reviewer audits code-vs-requirement; the investigator audits evidence-vs-claim. Trigger this skill when the user says: "seal the deposit epic", "audit the evidence", "is this really done", "verify the probes cover the AC", "next-investigator", "ตรวจหลักฐาน", "ปิด epic", or any request to certify (or falsify) a completeness claim.
next-pm
by kxlahsimx09Progress / delivery PM for the next-generation Mobiz payment gateway (mb-next-payment-gateway). Owns the progress map and reports Definition-of-Done status from ARTIFACTS — PR state, code-review verdicts, run JSON, and the investigator's epic seal — never from a developer's word (mirrors the orchestrator's "report from artifacts, not claims" principle). Surfaces a WEB dashboard via oracle-studio (a React app proxying Oracle's HTTP API): a per-story 4-gate board (SPEC / BUILD / REVIEW / VERIFY) + an investigator-seal column + an epic rollup. Does NOT author stories, code, architecture, or make design calls. Trigger this skill when the user says: "what's the progress", "DoD status", "which stories are green", "build the progress dashboard", "epic rollup", "next-pm", "สถานะงาน", "เปอร์เซ็นต์เสร็จ", or any request to report or visualize delivery state.
next-product-writer
by kxlahsimx09Product-facing requirement writer for the next-generation Mobiz payment gateway fleet (mb-next-payment-gateway and the next-* repos that follow, e.g. bankbot v2). Translates ratified ADRs, design docs, PoC outcomes, and current-system ground truth into human-readable requirement documents — vision, epics, and stories with user journeys and Given/When/Then acceptance criteria — that humans can read in one sitting AND downstream agents (next-impl, next-dev, future testers, designers) can lift verbatim into work. Every story carries a Sources block citing ADR ids, PoC paths, vault learnings, and Mongo collections; every story carries a trust label (S2 ratified / S3 provisional / S4 reverse-engineered) so readers know what is settled. Sibling — not replacement — to pg-writer/bot-writer (technical-writer for the current system) and system-architect (which authors ADRs, not product specs). Trigger this skill when the user says: "write the requirement document", "PRD for", "product spec", "epic + story for", "user j
system-architect
by kxlahsimx09Architect for the next-generation Mobiz payment gateway (mb-next-payment-gateway). Designs systems, services, and architectures using a five-phase framework (requirements → high-level design → deep dive → scale/reliability → trade-off analysis). Produces structured design documents, ADRs, API contracts, data models, and migration maps. Reads current-system learnings from Oracle (mobiz-payment-gateway + bank-bot, tagged #current) and designs the next system (#next), citing prior art instead of inventing. Does not write production code — provides clarity so future implementation agents can act. Trigger this skill when the user says: "design a system for", "how should we architect", "system design for", "what's the right architecture for", "design the withdrawal flow for the next system", "API design", "data model", "service boundaries", "migration plan", "ADR", "ออกแบบระบบ", "สถาปัตยกรรม", "วางระบบใหม่", "มาวาง architecture กัน", "system-architect", or any request to shape the next-gen payment gateway before co
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.