Directory
Skills and plugins
Search the directory and filter by grade or ecosystem. Live entries are A, B, C and D. No F grades are published yet. 15739 entries are listed. Only entries the scanner has graded are listed.
Grades are a hygiene check, not a security guarantee.
This grade is a hygiene screen, not proof of safety.
Showing 951 to 1000 of 1,212 Claude Code, Grade C entries. Page 20 of 25.
Listed A to Z by name.
No entries match that filter.
Grade C, One medium finding, Hexagon
INSTALL_REMOTE_SCRIPT
scripts/bootstrap.sh:7 # curl -fsSL https://raw.githubusercontent.com/Mexregkan/chatgpt-for-researchers/main/sChatgpt For Researchers (Skill)
Compatibility: Not stated.
This tool helps researchers organize long projects.
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
INSTALL_REMOTE_SCRIPT
install.sh:5 # curl -fsSL https://raw.githubusercontent.com/bd-loser/claude-code-termux/main/install.sh | bashClaude Code Termux (Skill)
Compatibility: Not stated.
This project allows you to use the Claude Code command line tool on your Android phone.
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
SECRETS_PATH_CREDENTIAL SECRETS_PATH_DOTENV
scripts/validate.py:1055 if name in {".netrc", ".npmrc", "credentials.json", "id_dsa", "id_ecdsa", "id_ed25519Claude Codex Skills Assembly (Skill)
Compatibility: Not stated.
Reusable agent skills, deterministic activation, safety hooks, and installers for Claude Code and OpenAI Codex, with an agentskills.io-compatible skill surface.
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
PERMISSION_SHELL_UNDECLARED SECRETS_PATH_UNCONFIRMED
plugins/changelog-keeper/scripts/hook.py:55 out = subprocess.run(Claude Plugins (Plugin)
Compatibility: Not stated.
Git workflow discipline for teams working in Claude Code.
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
PERMISSION_DELETE_FILES NETWORK_NO_FIXED_HOST NETWORK_UNNAMED_HOST +2 more
plugins/feature-loop/skills/feature-loop/scripts/log_run.py:44 proc = subprocess.run(["git", "-C", repo, "rev-Claude Skills (Plugin)
Compatibility: Not stated.
Skills personnels pour [Claude Code](https://claude.com/claude-code) , une famille cohérente qui couvre le cycle de vie complet d'une tâche de dev, du cadrage à la clôture.
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
PERMISSION_SHELL_UNDECLARED
src/codex_wake/adapters.py:36 result = subprocess.run(command, capture_output=True, text=True, timeout=10)Codex Wake (Skill)
Compatibility: Not stated.
<h1 align="center">Codex Wake</h1>
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
PERMISSION_DELETE_FILES NETWORK_NO_FIXED_HOST SECRETS_PATH_CREDENTIAL +1 more
appearance-service.js:387 fs.rmSync(temporary, { force: true })Deepseek Pa Dsh Desktop (Skill)
Compatibility: Not stated.
[English](README.md) | [简体中文](README.zh-CN.md)
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
PERMISSION_SHELL_UNDECLARED SECRETS_PATH_DOTENV
determify/ast_scanner.py:631 # DET-07: Subprocess calls running agent CLIs (e.g. subprocess.run(["opencode", "determify (Skill)
Compatibility: Not stated.
Set up (with user consent) and execute the Determify CLI to audit codebases for deterministic AI waste (date math, file checks, JSON/YAML parsing, formatting, or ungated SDK calls), interpret findings, and collaboratively refactor code.
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
SECRETS_PATH_UNCONFIRMED OBFUSCATION_MINIFIED_PAYLOAD
.gitignore:9 .codex/Dialogue Trace (Skill)
Compatibility: Not stated.
Record and resume research questions, evidence and experiments across Codex tasks in explicitly registered local workspaces.
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
PERMISSION_SHELL_UNDECLARED SECRETS_PATH_DOTENV
skills/docforge/_shared/runtime/common/python/repo_identity.py:54 result = subprocess.run(docforge (Skill)
Compatibility: Not stated.
Writes new repository documentation from source-code evidence , catalog-driven, provenance-tracked, and independently audited.
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
SECRETS_PATH_CREDENTIAL SECRETS_PATH_DOTENV
.gitignore:25 id_rsa*Doubao Geo Skills (Skill)
Compatibility: Not stated.
一套面向中文内容运营者、本地商家和 B2B 营销团队的 GEO（生成式引擎优化）Skill 工具包，把客户资料整理、选题、写稿、平台改写和豆包检验串成可复用流程，减少重复整理，让内容生产与复盘有据可查。
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
NETWORK_NO_FIXED_HOST OBFUSCATION_MINIFIED_PAYLOAD
lib/client.js:106 const response = await fetch(SETTINGS_ROUTE, body === void 0 ? { credentials: "Dsh Paddleocr Skills (Skill)
Compatibility: Not stated.
English | [简体中文](README.zh-CN.md)
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
NETWORK_NO_FIXED_HOST SECRETS_PATH_CREDENTIAL
skills/skillnotary/reference/capabilities.md:20 | `network` | `curl`, `wget`, `fetch()`, `axios`, `requests.*`Dsh Skillnotary (Skill)
Compatibility: Not stated.
**A DSH plugin that pins what your agent's skills are allowed to do , and tells you when that changes.**
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
PERMISSION_SHELL_UNDECLARED NETWORK_NO_FIXED_HOST
evals/results/generic-v0.2-codex-answers-r2/records/routing-positive-refactoring.json:89 "command": "/bin/zshEvolve Software Architecture (Skill)
Compatibility: Not stated.
An agent Skill for evidence-based software architecture guidance.
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
PERMISSION_DELETE_FILES INSTALL_LIFECYCLE_HOOK NETWORK_NO_FIXED_HOST
scripts/gen-skill-docs.ts:193 for (const file of obsolete) fs.rmSync(path.join(skillsDocDir, file));Flow Skills (Skill)
Compatibility: Not stated.
Portable, agent-agnostic workflow skills: plan, build, test, review, verify, orchestrate and ship.
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
SECRETS_PATH_DOTENV OBFUSCATION_MINIFIED_PAYLOAD
.gitignore:5 .envfluid (Skill)
Compatibility: Not stated.
<p align="center"> <a href="https://fluid.plrstudio.fr"> <img src="assets/banner.svg" alt="fluid, production-grade UI for AI agents, minus the slop" width="100%"> </a> </p>
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
NETWORK_NO_FIXED_HOST SECRETS_PATH_CREDENTIAL SECRETS_PATH_DOTENV
src/managed.ts:116 const response = await fetch(url, { signal: AbortSignal.timeout(120_000) });Google Workspace Agent Pack (Skill)
Compatibility: Not stated.
Private Google Workspace tooling for agents: a pinned managed `gws` CLI install, explicit human OAuth, offline/live diagnostics, portable Workspace skills, and optional MCP configuration templates.
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
NETWORK_NO_FIXED_HOST SECRETS_PATH_CREDENTIAL SECRETS_PATH_DOTENV
blueprint/hooks/safety-guard.sh:226 # curl/wget piped to shell (remote code execution)Gtm Os Kit (Skill)
Compatibility: Not stated.
<p align="center"> <img src="https://img.shields.io/badge/Claude_Code-Ready-blueviolet?style=for-the-badge" alt="Claude Code Ready" /> <img src="https://img.shields.io/badge/Skills-5_Included-orange?style=for-the-badge" alt="5 Skills" /> <img src="https://img.shields.io/badge/Plu
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
PERMISSION_SHELL_UNDECLARED INSTALL_UNPINNED SECRETS_PATH_DOTENV
examples/changelog-forge/scripts/generate_changelog.py:61 proc = subprocess.run(How To Create A Skill Tutorial (Skill)
Compatibility: Not stated.
<div align="center">
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
INSTALL_REMOTE_SCRIPT
install.sh:6 # curl -fsSL https://raw.githubusercontent.com/haidrrrry/humanize-ai-writing/main/install.sh | baHumanize Ai Writing (Skill)
Compatibility: Not stated.
Most AI writing services produce predictable text.
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
PERMISSION_SHELL_UNDECLARED PERMISSION_DELETE_FILES NETWORK_NO_FIXED_HOST
skills/ib-suite/scripts/bootstrap.py:168 shutil.rmtree(stale, ignore_errors=True)Ib Suite (Skill)
Compatibility: Not stated.
**English** | [简体中文](README.zh-CN.md)
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
SECRETS_PATH_CREDENTIAL SECRETS_PATH_DOTENV
bin/jj-land:182 .env|.env.*|*.pem|*.key|*.p12|*.pfx|*.keystore|id_rsa*|id_dsa*|id_ecdsa*|id_ed25519*) note="seJj Agentic Workflow (Skill)
Compatibility: Not stated.
A practical methodology for running coding agents , including several in parallel , on [Jujutsu (jj)](https://github.com/jj-vcs/jj) repositories, without them destroying each other's work.
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
PERMISSION_SHELL_UNDECLARED PERMISSION_DELETE_FILES NETWORK_NO_FIXED_HOST
junyi-ai-weekly-review/scripts/collect.mjs:64 else if (entry.isFile() && /^chunk_\d+\.md$/.test(entry.name)) fJunyi Skills (Skill)
Compatibility: Not stated.
**Junyi’s AI Collaboration Skills**
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
NETWORK_HOST_FROM_INPUT
kevin-xhs-minitool-publisher/scripts/fetch_official_skill.py:39 with urllib.request.urlopen(request, timeout=6Kevinsu Skills (Skill)
Compatibility: Not stated.
<div align="center">
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
PERMISSION_SHELL_UNDECLARED PERMISSION_DELETE_FILES
starter-kit/tools/scrub_outbound.py:112 out = subprocess.run(["git", "diff", "--cached", "--name-only", "--difLucid Sheep Skill (Skill)
Compatibility: Not stated.
>
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
NETWORK_NO_FIXED_HOST SECRETS_PATH_DOTENV OBFUSCATION_MINIFIED_PAYLOAD
skills/character-sprites/assets/preview.html:13 const response=await fetch('atlas.json');if(!response.ok)throwNoodle Skills (Skill)
Compatibility: Not stated.
**Ready-to-install agent skills, each wrapping a [nanoodle](https://nanoodle.com) visual AI workflow.** Drop a folder into your agent's skills directory and Claude Code (or any agent that reads markdown playbooks and runs shell commands) can generate posters, restyle photos, sing
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
NETWORK_NO_FIXED_HOST SECRETS_PATH_DOTENV OBFUSCATION_MINIFIED_PAYLOAD
core.py:95 with urllib.request.urlopen(urllib.request.Request(self.url, body, headers), timeout=180) as responNovel Graph Lab (Skill)
Compatibility: Not stated.
<p align="center"><img src="docs/images/mark.svg" width="64" height="64" alt="Novel Graph Lab"></p>
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
PERMISSION_SHELL_UNDECLARED
one-pager/scripts/export.py:50 subprocess.run(One Page Doc Skills (Skill)
Compatibility: Not stated.
An agent skill for compressing anything , a book, a 60-page report, meeting notes, a proposal , into exactly one A4 page that argues like an Amazon memo and looks like a magazine's art desk made it.
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
NETWORK_HOST_FROM_INPUT
meritz_studio/core/client.py:237 r = requests.post(Open Api Studio (Skill)
Compatibility: Not stated.
> 현재 베타 서비스 기간입니다.
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
NETWORK_HOST_FROM_INPUT
docs/skills/backend-suite-orchestrator/SKILL.md:229 with urllib.request.urlopen(url, timeout=10) as resp:Opencode Artes (Skill)
Compatibility: Not stated.
<div align="center">
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
PERMISSION_SHELL_UNDECLARED PERMISSION_DELETE_FILES
office-docx-editor/render_preview.py:58 return subprocess.run(cmd, capture_output=True, text=True, timeout=timOpencode Office Skills (Skill)
Compatibility: Not stated.
符合 [OpenCode Agent Skills](https://opencode.ai/docs/skills/) 规范的 Office 文档自动化编辑技能集，覆盖 Word / PowerPoint / Excel 三大格式。每个 skill 都是独立目录，包含 `SKILL.md`（agent 使用的说明文档）和可直接执行的 CLI 脚本。
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
PERMISSION_SHELL_UNDECLARED PERMISSION_DELETE_FILES
benchmark/build.py:653 shutil.rmtree(corpus_dir)Orange Check (Skill)
Compatibility: Not stated.
> 이미 써 놓은 문서의 인용을 하나씩 열어 봅니다.
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
INSTALL_LIFECYCLE_HOOK NETWORK_NO_FIXED_HOST SECRETS_PATH_UNCONFIRMED
package.json:55 prepareP Backlog (Skill)
Compatibility: Not stated.
**[Русская версия](README.ru.md)**
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
PERMISSION_SHELL_UNDECLARED PERMISSION_DELETE_FILES SECRETS_PATH_DOTENV
evals/_harness/run_skill.py:75 proc = subprocess.run(Partnerships Skills (Skill)
Compatibility: Not stated.
Skills for partner ecosystem operations , affiliate programs, influencer partnerships, channel alliances, joint GTM, partner enablement, and ecosystem measurement.
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
PERMISSION_DELETE_FILES NETWORK_NO_FIXED_HOST NETWORK_UNNAMED_HOST +2 more
scripts/figma_mcp.py:35 self.process = subprocess.Popen(command, stdin=subprocess.PIPE, stdout=subprocess.PIPEPptedit (Skill)
Compatibility: Not stated.
创建、修改、调试和验收 H5、HTML 单页、落地页与网页 Deck。当任务需要修改真实网页源码并在浏览器中验证时使用；不适用于没有 HTML 运行时的原生演示文件。
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
NETWORK_NO_FIXED_HOST SECRETS_PATH_DOTENV OBFUSCATION_MINIFIED_PAYLOAD
packages/claude-code-plugin/skills/remembrancer/references/attestation-rest.md:380 const response = await fetcRemembrance Skills (Skill)
Compatibility: Not stated.
Discover and directly invoke authorized Remembrance skills in VS Code agent mode, then close the outcome and feedback loop.
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
PERMISSION_SHELL_UNDECLARED SECRETS_PATH_DOTENV
tools/test_fitness_visuals.py:108 result = subprocess.run([os.environ.get("CODEX_PRIMARY_RUNTIME_NODE", "node"Rowan Fitness Skill (Skill)
Compatibility: Not stated.
Rowan is a free, open-source **AI fitness coaching skill for Claude and Codex**, built for intermediate and advanced trainees.
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
PERMISSION_SHELL_UNDECLARED PERMISSION_DELETE_FILES NETWORK_NO_FIXED_HOST +1 more
evals/_harness/run_skill.py:75 proc = subprocess.run(Sales Skills (Skill)
Compatibility: Not stated.
Skills for running a sales org and closing the deals in it , market sizing, ICP, quota and comp design, cold outbound, discovery, MEDDPICC, negotiation, deal strategy, and post-sale motion.
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
SECRETS_PATH_UNCONFIRMED OBFUSCATION_MINIFIED_PAYLOAD
.gitignore:4 .DS_StoreSandKit (Skill)
Compatibility: Not stated.
<div align="center">
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
PERMISSION_DELETE_FILES NETWORK_NO_FIXED_HOST SECRETS_PATH_CREDENTIAL +1 more
cli/lib/commands/rules.js:89 if (created) fs.rmSync(targetAbs, { force: true });sasu (Skill)
Compatibility: Not stated.
<p align="center"><img src="assets/mascot.png" width="260" alt="sasu mascot"/></p>
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
INSTALL_REMOTE_SCRIPT
sessions/inbox/pyros-2026-08-10-185a41c4.md:41 curl -fsSL https://getpyros.com/install.sh | bashSecondBrain (Skill)
Compatibility: Not stated.
<img width="1448" height="1086" alt="image" src="https://github.com/user-attachments/assets/61e77098-42d9-43cc-8f03-3eb335066d8d" />
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
NETWORK_NO_FIXED_HOST SECRETS_PATH_CREDENTIAL SECRETS_PATH_DOTENV
tool/extract_oscal/extract_oscal.py:208 # The site's CDN returns 403 to the default curl and python-requests USg Im8 Reform Skill (Skill)
Compatibility: Not stated.
This repo turns Singapore Government Instruction Manual for ICT&SS (previously known as **IM8**) into an **agent skill**.
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
PERMISSION_SHELL_UNDECLARED
scripts/validate_skill.py:16 subprocess.run(command, cwd=ROOT, check=True)Shao Alan Operating System (Skill)
Compatibility: Not stated.
|
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
NETWORK_NO_FIXED_HOST SECRETS_PATH_CREDENTIAL SECRETS_PATH_DOTENV
skills/agentpub-publish/SKILL.md:35 curl -sX POST https://agentpub.io/api/v1/publish \skill (Skill)
Compatibility: Not stated.
Turn agent work into owned, reviewable web pages.
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
NETWORK_NO_FIXED_HOST SECRETS_PATH_UNCONFIRMED OBFUSCATION_MINIFIED_PAYLOAD
engine.py:82 'supply_chain': 'Downloads and executes remote code (curl piped to a shell, unpinned installers,Skill Dash (Skill)
Compatibility: Not stated.
Judge a tree of Claude Code and Codex skills with [Jev](https://typesafe.ai), TypeSafe's typed-judgment model.
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
PERMISSION_DELETE_FILES SECRETS_PATH_DOTENV OBFUSCATION_MINIFIED_PAYLOAD
evals/run_linux_backend_check.py:13 files = subprocess.check_output(['git', 'ls-files'], cwd=ROOT, text=True).Skill Forge (Skill)
Compatibility: Not stated.
Audit, validate, pressure-test, and grade OpenAI or portable Agent Skills from ZIPs, folders, or SKILL.md drafts; diagnose triggering, suggest fixes, and assess release readiness.
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
PERMISSION_DELETE_FILES NETWORK_NO_FIXED_HOST NETWORK_UNNAMED_HOST +2 more
src/runtime-safety.test.ts:22 if(e.clean) fs.rmSync(e.cleanPath||'draft.txt',{force:true});skillhook (Skill)
Compatibility: Not stated.
Turn this machine into a permanent, secure webhook endpoint that runs Agent Skills with Claude Code or Codex, and run your whole Skillhook Cloud organisation from the agent.
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
NETWORK_HOST_FROM_INPUT
templates/admin-dashboard/services/web/app/resource-table.js:28 const res = await fetch(`/api/${resource}`, {skills (Skill)
Compatibility: Not stated.
Build and run apps on CloudGrid from Claude Code: brainstorming, planning, templates, and one-step plugs to a live URL - on a live runtime that provides managed databases, cache, disk, and AI (no API keys) out of the box
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
PERMISSION_DELETE_FILES SECRETS_PATH_UNCONFIRMED OBFUSCATION_MINIFIED_PAYLOAD
scripts/validate.py:172 proc = subprocess.run(skills (Skill)
Compatibility: Not stated.
jihkyuo's personal agent skills: design-variant-rounds (parallel design variants with round engines), socratic dialogue, orchestration loop, Orca handoff
This grade is a hygiene screen, not proof of safety.
Grade C, One medium finding, Hexagon
AGENT_IGNORE_INSTRUCTIONS
skills/skill-curator/references/security-screen.md:36 - **Instructions to disable safety behavior**, ignore prskills (Skill)
Compatibility: Not stated.
Three skills for Claude Code, Codex, Cursor, Grok and the other agents the [`skills` CLI](https://github.com/vercel-labs/skills) installs into: one for prose, one for UI review, one for keeping a skill library honest.
This grade is a hygiene screen, not proof of safety.
