Claude Marketplaces
SkillsMCP serversMarketplacesDigestTools
Submit

This week in Claude

Every Monday: Claude Code, Agent SDK, MCP, and the Anthropic platform moves worth your time.

Skills by category
Frontend DevelopmentBackend & APIsTesting & QASecurityDevOps & CI/CDGit & Pull RequestsDocumentationCode Review & QualityAI & Agent BuildingSkill Development
MCP servers by category
Sales & MarketingWeb & Browser AutomationDatabasesAI & LLM ToolsCloud & InfrastructureCommunication & MessagingDeveloper ToolsDesign & CreativeDocuments & KnowledgeSearch & Web Crawling
Marketplaces by category
AI Agents & OrchestrationLLM IntegrationDevelopment ToolsFrontend & UIBackend & APIsDatabasesTesting & Code QualityDevOps & CloudSecurity & ComplianceGit & Version Control
Claude Marketplaces

An independent directory of skills, plugins and MCP servers for Claude Code. Not affiliated with Anthropic.

Browse

  • Skills
  • MCP servers
  • Marketplaces
  • Weekly digest
  • Skill index
  • MCP index
  • Marketplace index

Site

  • About
  • Tools
  • Submit
  • Feedback
  • Privacy
  • llms.txt
  • Plugins reference

Maker

  • Snug
  • Agent Caffeine
  • knobs
  • Wax
  • Vectoys
  • mertduzgun.com
Crafted at
Skills/AI & Agent Building
telagod avatar

Cultivating Skills

telagod/code-abyss

Turns repeated workflows into reusable skills through a three-tier publishing funnel: local for private experiments, project for team sharing, and community for upstream contribution. The distill mode kicks in when Claude notices you've done something three times. The improve mode patches existing skills that fall short. Everything goes through a safety scan that blocks hardcoded secrets, overprivileged tool access, and injection patterns before it hits disk. Default is local-only, which keeps the noise down. If you find yourself copy-pasting the same prompt structure or command chain across sessions, this is the crystallization layer. Ships with a Node-based forge script that handles scaffolding, linting, and tier promotion.

GitHub stars
224
last updated
Jun 10, 2026
View on GitHubSave

Install

$ npx -y skills add telagod/code-abyss --skill cultivating-skills --agent claude-code

Installs into .claude/skills of the current project.

Files

炼制神通 · cultivating-skills

框架之外仍有道。重复三次即沉淀,沉淀必经安全脊柱,发布必走漏斗。

三种触发模式

模式触发入口
distillAgent 在会话中识别"魔尊已重复 N 次同类操作"主动提议→显式确认
create用户显式 /cultivate-skill <topic>脚手架交互
improve用户指现有 skill 不足diff + 增量补丁

三级沉淀漏斗

L0 本地私有  ~/.claude/skills/local/         → 不入路由、显式调用、不审
L1 项目私有  <repo>/.claude/skills/          → 入 git、团队共享、强 lint
L2 社区贡献  GitHub Issue/PR + safety_scan   → 进入 upstream 候选池

默认 L0——魔尊主动升级,避免污染。每升一级,门槛递增。

何时使用

场景使用理由
同类操作 ≥ 3 次(手动构造的 prompt、命令链、修复模板)✅ distill边际收益最高
现有 skill 缺失关键场景 / 表述模糊✅ improve直接打补丁
用户提出新工程化方法论✅ create显式生成骨架

何时不使用

  • ❌ 单次性脚本 / 一次性查询——沉淀价值低,徒增噪音
  • ❌ 复制粘贴现有 skill 内容——用 aliases 或 reference 链接即可
  • ❌ 项目特定状态(API 端点、内部约定)——写进项目 CLAUDE.md
  • ❌ 重复内容含 secret / 私密细节——永不沉淀

安全脊柱(默认拒绝)

新 skill 落盘前必须通过 safety_scan,命中任一项即阻断:

  1. frontmatter 合规:name kebab-case 唯一、description ≥ 40 字、user-invocable 显式声明
  2. 工具最小权限:allowed-tools 默认仅 Read;扩权(Bash/Write/Edit/WebFetch)必须有理由说明
  3. 无硬编码 secret(复用 analyzing-security 规则集)
  4. 无危险默认模板、无 prompt injection 反模式、scripts 单入口、引用文件存在

详细规则矩阵见 references/safety-review.md。

工作流速查

你想做的走哪卷
从会话提炼新 skillcreation-workflow.md
改进现有 skill 的某个场景improvement-workflow.md
识别"该沉淀了"的信号distillation-patterns.md
通过安全审查safety-review.md
升级到项目内 / 提交到社区publishing-guide.md

使用

# 脚手架:默认 L0 本地,可改 --tier project / community
node scripts/skill_forge.js init <slug> [--tier local|project|community]

# 本地校验(frontmatter + 引用 + 工具白名单)
node scripts/skill_forge.js lint <skill-dir>

# 安全扫描(落盘前必跑)
node scripts/skill_forge.js scan <skill-dir>

# 改进模式(生成 diff 草案)
node scripts/skill_forge.js improve <existing-skill-dir>

# 升级 tier
node scripts/skill_forge.js promote <skill-dir> --to project

收口

落盘前必须:scan 通过 + lint 通过 + 魔尊 review diff。 社区提交:复用 submission portal,本 skill 只负责生成 payload,不重造提交流程。

参见姊妹 skill cultivating-personas——专司人格沉淀。

SKILL.md on GitHub

Sponsors

SnugThe clipboard manager you and your agent can use, safelyAgent CaffeineClose your lid, come back to finished workWaxBlock anything with supervision, without erasing your iPhoneOpen StepsSponsoredFree and open-source skills that make AI coding agents easier to understand, verify, and control.

Details

Repository
github.com/telagod/code-abyss
Categories
AI & Agent Building, Release Management
License
MIT
First seen
Jun 11, 2026

More from telagod/code-abyss

All 30
  • Defending Applications
  • Designing Architectures
  • Designing Hardware Products
  • Detecting And Responding
  • Developing Mobile Apps
  • Developing Software
  • Engineering Data Pipelines
  • Generating Docs
  • Indexing Code
  • Operating Kicad Eda
  • Orchestrating Adversarial Reviews
  • Processing Docx
  • Processing Pdfs
  • Provisioning Infrastructure
  • Reducing Aigc Detection
  • Securing Cloud And Supply Chain
  • Securing Systems
  • Shipping Changes
  • Verifying Modules
  • Analyzing Changes
  • Analyzing Security
  • Analyzing Spreadsheets
  • Applying Ui Design System
  • Architecting Security

Recommended

More AI & Agent Building
1telagod avatardesigning-hardware-productstelagod/code-abyssAI-automated hardware product pipeline: from requirements to manufacturable Gerber + compiled firmware + cross-platform app. Orchestrates system design, ESP-IDF0installs
2telagod avatarindexing-codetelagod/code-abyssCode relationship graph and temporal intelligence via abyss CLI. Provides caller tracing, impact analysis, hotspot detection, and file-level context gathering. 0installs
3telagod avataroperating-kicad-edatelagod/code-abyssKiCad EDA orchestration via kicad-mcp MCP server. Routes 17 tools for schematic creation, PCB layout, autorouting, DRC, and Gerber export. Enforces serialized P0installs
4ThinkInAIXYZ avataradd-providerThinkInAIXYZ/deepchatAdd a DeepChat LLM provider through explicit reviewed source changes. Use when a developer asks Codex to add a provider, provider profile, upstream provider con0installs
5thinkinaixyz avatarcua-driverthinkinaixyz/deepchatDrive native macOS apps through DeepChat's plugin-provided Computer Use MCP tools. Use when the user asks to operate, inspect, automate, or perform a GUI task i0installs
6thinkinaixyz avatarfeishu-toolsthinkinaixyz/deepchatUse the Feishu/Lark plugin MCP tools for Feishu documents, spreadsheets, knowledge content, and other matching workspace operations.0installs