Agent skills that run json, that ask for Edit, in other format

Skills declare what they need inconsistently: 58.8% name the languages they run, 11.2% list the tools they ask for, and 4.6% give themselves a category. Filtering narrows to the ones that said so, never to the ones that are.

1. opencli-browserjackwener/OpenCLI

   Use when an agent needs to drive a real Chrome window via opencli — inspect a page, fill forms, click through logged-in flows, or extract data ad-hoc. Covers the selector-first tar

   otherbashjsonBash(opencli:*)ReadEdit
2. auto-paper-improvement-loopwanshuiyin/Auto-claude-code-research-in-sleep

   Autonomously improve a generated paper via GPT-6-Astra xhigh review → implement fixes → recompile, for 2 rounds. Use when user says "改论文", "improve paper", "论文润色循环", "auto improve"

   otherbashjsonmarkdownyamlBash(*)ReadWrite+1 files
3. vast-gpuwanshuiyin/Auto-claude-code-research-in-sleep

   Rent, manage, and destroy GPU instances on vast.ai. Use when user says "rent gpu", "vast.ai", "rent a server", "cloud gpu", or needs on-demand GPU without owning hardware.

   otherbashjsonmarkdownBash(*)ReadWrite
4. figure-specwanshuiyin/Auto-claude-code-research-in-sleep

   Generate deterministic publication-quality architecture, workflow, and pipeline diagrams from structured JSON (FigureSpec) into editable SVG. Use when user says "架构图", "workflow 图"

   otherbashjsonBash(*)ReadWrite+1 files
5. paper-slideswanshuiyin/Auto-claude-code-research-in-sleep

   Generate conference presentation slides (beamer LaTeX → PDF + editable PPTX) from a compiled paper, with speaker notes and full talk script. Use when user says "做PPT", "做幻灯片", "mak

   otherbashjsonlatexmarkdownBash(*)ReadWrite
6. experiment-auditwanshuiyin/Auto-claude-code-research-in-sleep

   Audit experiment integrity before claiming results. Uses cross-model review (external reviewer backend) to check for fake ground truth, score normalization fraud, phantom results,

   otherjsonmarkdownBash(*)ReadWrite+1 files
7. paper-writingwanshuiyin/Auto-claude-code-research-in-sleep

   Workflow 3: Full paper writing pipeline that goes from a narrative report to a polished, submission-ready PDF. Use when user says "写论文全流程", "write paper pipeline", "从报告到PDF", "pape

   otherbashjsonmarkdownBash(*)ReadWrite+5 files
8. grant-proposalwanshuiyin/Auto-claude-code-research-in-sleep

   Draft a structured grant proposal from research ideas and literature. Supports KAKENHI (Japan), NSF (US), NSFC (China, including 面上/青年/优青/杰青/海外优青/重点), ERC (EU), DFG (Germany), SNSF

   otherbashjsonmarkdownBash(*)ReadWrite+3 files
9. research-refinewanshuiyin/Auto-claude-code-research-in-sleep

   Turn a vague research direction into a problem-anchored, elegant, frontier-aware, implementation-oriented method plan via iterative GPT-6-Astra review. Use when the user says "refi

   otherjsonmarkdownBash(*)ReadWrite+3 files
10. citation-auditwanshuiyin/Auto-claude-code-research-in-sleep

    Zero-context verification that every bibliographic entry in the paper is real, correctly attributed, and used in a context the cited paper actually supports — catching hallucinated

    otherbashjsonmarkdownBash(*)ReadGrep+1 files
11. patent-pipelinewanshuiyin/Auto-claude-code-research-in-sleep

    Full patent drafting pipeline from invention description to jurisdiction-formatted filing documents. Supports CN (CNIPA), US (USPTO), EP (EPO). Supports invention patents and utili

    otherjsonmarkdownBash(*)ReadWrite
12. interview-cheatsheetwanshuiyin/Auto-claude-code-research-in-sleep

    Generate a long-form Chinese interview-prep cheat sheet on a specific ML/LLM topic — formulas with derivations, from-scratch PyTorch code, comparison tables, and 25 高频面试题 (L1 必会 /

    otherbashjsonBash(*)ReadWrite
13. auto-review-loopwanshuiyin/Auto-claude-code-research-in-sleep

    Autonomous multi-round research review loop. In Copilot CLI it defaults to the native complementary rubber-duck subagent with host-event model evidence; elsewhere it uses Codex, wh

    otherbashjsonjsonlmarkdownBash(*)ReadGrep+4 files
14. dse-loopwanshuiyin/Auto-claude-code-research-in-sleep

    Autonomous design space exploration loop for computer architecture and EDA. Runs a program, analyzes results, tunes parameters, and iterates until objective is met or timeout. Use

    otherjsonmarkdownBash(*)ReadGrep+1 files
15. proof-checkerwanshuiyin/Auto-claude-code-research-in-sleep

    Rigorous mathematical proof verification and fixing workflow. Reads a LaTeX proof, identifies gaps via cross-model review (external reviewer backend, ultra reasoning), fixes each g

    otherbashjsonmarkdownBash(*)ReadGrep+3 files
16. kill-argumentwanshuiyin/Auto-claude-code-research-in-sleep

    Two-thread adversarial review: a fresh reviewer constructs the strongest 200-word rejection memo, then a second fresh reviewer defends the paper point-by-point and surfaces still-u

    otherbashjsonmarkdownBash(*)ReadWrite+2 files
17. auto-review-loop-minimaxwanshuiyin/Auto-claude-code-research-in-sleep

    Autonomous multi-round research review loop using MiniMax API. Use when you want to use MiniMax instead of Codex MCP for external review. Trigger with "auto review loop minimax" or

    otherbashjsonmarkdownBash(*)ReadGrep+4 files
18. auto-review-loop-llmwanshuiyin/Auto-claude-code-research-in-sleep

    Autonomous research review loop using any OpenAI-compatible LLM API. Configure via llm-chat MCP server or environment variables. Trigger with "auto review loop llm" or "llm review"

    otherbashjsonmarkdownBash(*)ReadGrep+4 files
19. paper-claim-auditwanshuiyin/Auto-claude-code-research-in-sleep

    Zero-context verification that every number, comparison, and scope claim in the paper matches raw result files. Uses a fresh cross-model reviewer with NO prior context to prevent c

    otherjsonmarkdownBash(*)ReadWrite+1 files
20. slides-polishwanshuiyin/Auto-claude-code-research-in-sleep

    Per-page Codex review + targeted python-pptx / Beamer fixes for academic talk slides. Use AFTER /paper-slides (or any externally generated PPTX/Beamer) when the deck looks 'mostly

    otherjsonlatexpythonBash(*)ReadWrite
21. aris-grant-proposalOpenLAIR/dr-claw

    Draft a structured grant proposal from research ideas and literature. Supports KAKENHI (Japan), NSF (US), NSFC (China, including 面上/青年/优青/杰青/海外优青/重点), ERC (EU), DFG (Germany), SNSF

    otherjsonmarkdownBash(*)ReadWrite
22. aris-auto-paper-improvement-loopOpenLAIR/dr-claw

    Autonomously improve a generated paper via GPT-5.4 xhigh review → implement fixes → recompile, for 2 rounds. Use when user says "改论文", "improve paper", "论文润色循环", "auto improve", or

    otherbashjsonmarkdownBash(*)ReadWrite
23. aris-dse-loopOpenLAIR/dr-claw

    Autonomous design space exploration loop for computer architecture and EDA. Runs a program, analyzes results, tunes parameters, and iterates until objective is met or timeout. Use

    otherjsonmarkdownBash(*)ReadGrep
24. aris-paper-posterOpenLAIR/dr-claw

    Generate a conference poster (article + tcbposter LaTeX → A0/A1 PDF + editable PPTX + SVG) from a compiled paper. Use when user says "做海报", "制作海报", "conference poster", "make poste

    otherbashjsonlatexmarkdownBash(*)ReadWrite
25. aris-paper-slidesOpenLAIR/dr-claw

    Generate conference presentation slides (beamer LaTeX → PDF + editable PPTX) from a compiled paper, with speaker notes and full talk script. Use when user says "做PPT", "做幻灯片", "mak

    otherbashjsonlatexmarkdownBash(*)ReadWrite
26. inno-grant-proposalOpenLAIR/dr-claw

    Help professors and researchers write, revise, adapt, and polish grant proposals for US agencies (NSF, NIH, DOE, DARPA, NASA) and Chinese agencies (NSFC 国自然). Use this skill whenev

    otherbashjsonReadWriteEdit
27. aris-research-refineOpenLAIR/dr-claw

    Turn a vague research direction into a problem-anchored, elegant, frontier-aware, implementation-oriented method plan via iterative GPT-5.4 review. Use when the user says "refine m

    otherjsonmarkdownBash(*)ReadWrite
28. aris-vast-gpuOpenLAIR/dr-claw

    Rent, manage, and destroy GPU instances on vast.ai. Use when user says "rent gpu", "vast.ai", "rent a server", "cloud gpu", or needs on-demand GPU without owning hardware.

    otherbashjsonmarkdownBash(*)ReadWrite
29. aris-auto-review-loopOpenLAIR/dr-claw

    Autonomous multi-round research review loop. Repeatedly reviews via Codex MCP, implements fixes, and re-reviews until positive assessment or max rounds reached. Use when user says

    otherbashjsonmarkdownBash(*)ReadGrep
30. agent-architecture-audittempletongroup/radiant

    Full-stack diagnostic for agent and LLM applications. Audits the 12-layer agent stack for wrapper regression, memory pollution, tool discipline failures, hidden repair loops, and r

    otherbashjsonReadWriteEdit
31. skill-creatorflonat/flonat-research

    Create, revise, and evaluate reusable AI workflow skills, including trigger-quality tests. Use when authoring a new skill, repairing an existing skill, or measuring whether its met

    otherbashjsonmarkdownReadWriteEdit+3 files
32. init-project-lightflonat/flonat-research

    Bootstrap a lightweight project with minimal guidance, context, and repository structure. Use when a small non-research project needs durable AI collaboration without the full rese

    otherjsonBash(mkdir*)Bash(ls*)Bash(touch*)
33. ideasflonat/flonat-research

    Capture and integrate improvement ideas for the shared Claude Code and Codex infrastructure. Use when recording an infrastructure enhancement for later review rather than implement

    otherjsonmarkdownReadWriteEdit
34. latex-health-checkflonat/flonat-research

    Compile all LaTeX projects and report cross-project build consistency. Use when checking whether a collection of papers builds cleanly. Not for rendered visual inspection after a c

    otherbashjsonReadEditWrite+2 files
35. citation-managementbeita6969/ScienceClaw

    Comprehensive citation management for academic research. Search Google Scholar and PubMed for papers, extract accurate metadata, validate citations, and generate properly formatted

    otherbashbibtexjsonReadWriteEdit
36. loom-api-designcosmix/loom

    Designs RESTful APIs, GraphQL schemas, and RPC interfaces for consistency, usability, and scalability.

    othergraphqljsonprotobufyamlReadGrepGlob
37. loom-grafanacosmix/loom

    Observability visualization with Grafana and the LGTM stack.

    otherjsonlogqlpromqltraceqlReadGrepGlob
38. api-developerVinix24/vnx-orchestration

    API developer specializing in clean, well-documented REST APIs

    otherjsonReadWriteEdit
39. caddy-mutewrg32786/aigent-os

    Mute a Caddy hint class for a window. Defaults to 4h auto-expiry. Classes — memory, context, body, routing, all.

    otherjsonReadWriteEdit
40. create-bossnicepkg/boss-skill

    Distill your boss into an AI Skill — PUA detection (10 corp flavors), counterattack coach, evidence collector, labor law lookup, cake-promise BS meter, resignation script, replacem

    otherbashjsonmarkdownReadWriteEdit
41. mir-cloud-cloudflareanantbhandarkar/make-it-right

    Make It Right (Cloudflare module). The reliability and cost review layer for a Cloudflare deployment, loaded after the provider is chosen - not a how-to for Workers. Covers: the is

    otherjsonReadWriteEdit
42. mir-cloud-awsanantbhandarkar/make-it-right

    Make It Right (AWS module). AWS identity and service mechanics, loaded after the provider is chosen. Covers: IAM evaluation order and where a permissions boundary does not bind; ia

    otherhcljsonyamlReadWriteEdit
43. mir-cloud-azureanantbhandarkar/make-it-right

    Make It Right (Azure module). Microsoft Entra identity and Azure service mechanics, loaded after the provider is chosen. Covers: app registration vs. service principal vs. managed

    otherhcljsonReadWriteEdit
44. mir-backend-nodeanantbhandarkar/make-it-right

    Make It Right (Node.js runtime tier). V8/Node 22–26 runtime reliability footguns that are shared across EVERY Node backend framework (Express, Fastify, NestJS, Hapi, Koa) — distinc

    otherdockerinijavascriptjsonReadWriteEdit
45. self-improving-agentzhouguoqing/QianYuan.AIAgenticFramework

    A universal self-improving agent that learns from ALL skill experiences. Uses multi-memory architecture (semantic + episodic + working) to continuously evolve the codebase. Auto-tr

    otherjsonmarkdownyamlReadWriteEdit
46. scammernaxiaoduo/Scammer.skill

    骗子.skill — 把骗局套路蒸馏成 AI，检测可疑消息，定位当前阶段，预判下一步。Distill scam patterns into AI, detect suspicious messages, locate current phase, predict next move.

    otherbashjsonReadWriteEdit
47. ag-criar-skillandregusman-raiz/a-gusman-claude

    Cria, melhora e avalia skills com evals quantitativos. Use para criar/atualizar/otimizar skill, rodar evals, benchmark de performance, ou otimizar description para triggering.

    otherbashjsonReadWriteEdit
48. ag-13-limpar-codigoandregusman-raiz/a-gusman-claude

    Limpeza de dead code com Knip+AST+bundle, confidence tiers e PRs atomicos. Use para tech debt, cleanup, bundle e --target=harness.

    otherbashjsontypescriptReadWriteEdit
49. ag-0-orquestradorandregusman-raiz/a-gusman-claude

    Entry point: classifica pedidos, roteia skills/agents/plugins, monitora execucao e sugere combinacoes canonicas para tarefas multi-etapa.

    otherbashjsonyamlReadGlobGrep
50. ag-5-documentosandregusman-raiz/a-gusman-claude

    Documentacao: Office (PPTX/DOCX/XLSX/PDF), README, API, diagramas, specs, changelog, data dictionary e CSV; executive para decks.

    otherbashjsonpythontypescriptReadWriteEdit
51. unikit-docsNintendaDev/unikit-ai

    Generate and maintain the project's TECHNICAL documentation from its codebase — scans the project structure, tech stack, and module boundaries, then writes a lean README landing pa

    otherjsonmarkdownReadWriteEdit
52. unikit-evolveNintendaDev/unikit-ai

    Learn project rules from accumulated fix patches in .unikit/code/patches/. Analyzes past mistakes, extracts prevention points, classifies each as a coding rule (→ RULES.md) or a pe

    otherjsonmarkdownReadWriteEdit
53. repo-scanhaibindev/repo-scan

    对指定项目源码目录执行全面资产审计，生成《全网模块与源码资产审计详细清单》。当用户要求"审计源码"、"盘点代码资产"、"生成清单"时自动触发。

    otherbashjsonmarkdownBashReadGlob+1 files

Browse all 69,865 skills →200 at a time, grouped by owner. The list above is the skills with the highest standing; this is every one of them.

The whole corpus is one call away through the search API, or ask a question of it on the index itself.