← 返回 Skills 市场
levineam

Context Engineering (Koylan)

作者 levineam · GitHub ↗ · v1.0.0
cross-platform ⚠ suspicious
413
总下载
0
收藏
1
当前安装
1
版本数
在 OpenClaw 中安装
/install levineam-context-engineering
功能描述
Comprehensive context engineering guidance for AI agent systems. Routes to specialized sub-skills for production agent work. Use when the user asks to: "opti...
使用说明 (SKILL.md)

Context Engineering Skill

Routes to the appropriate sub-skill from Koylan's Agent-Skills-for-Context-Engineering collection, located at:

references/context-engineering-skills/skills/\x3Csub-skill>/SKILL.md

Setup

No setup required. This skill uses GitHub raw URLs as the primary method to fetch sub-skill documentation.

Sub-skill SKILL.md files are loaded directly from GitHub:

https://raw.githubusercontent.com/muratcankoylan/Agent-Skills-for-Context-Engineering/main/skills/\x3Csub-skill>/SKILL.md

Example: For context-optimization, use:

https://raw.githubusercontent.com/muratcankoylan/Agent-Skills-for-Context-Engineering/main/skills/context-optimization/SKILL.md

Optional Enhancement (Offline Use): For offline access or faster local reads, you can optionally initialize the git submodule:

git submodule update --init

This will populate the references/context-engineering-skills/ directory locally. When the local directory exists, the skill loader will prefer local files over GitHub fetches.

Sub-Skill Routing Table

Match the user's task to one sub-skill and read its SKILL.md before proceeding. Load additional sub-skills only if the task spans multiple domains.

Task / trigger phrase Sub-skill directory
"optimize context", reduce token costs, KV-cache, context budgeting context-optimization
"context compression", summarize history, compaction, long sessions context-compression
"design multi-agent", supervisor pattern, swarm architecture, agent handoffs multi-agent-patterns
"implement memory", memory frameworks, Mem0, Zep, Letta, vector RAG memory-systems
"offload context to files", filesystem memory, scratch pads, dynamic discovery filesystem-context
"debug agent failure", lost-in-middle, context poisoning, degradation patterns context-degradation
"build agent tools", tool descriptions, MCP tools, tool consolidation tool-design
"evaluate agent", LLM-as-judge, test framework, quality gates evaluation
Advanced evaluation: rubrics, pairwise comparison, position bias advanced-evaluation
"understand context", context windows, attention mechanics, fundamentals context-fundamentals
"build background agent", sandboxed execution, hosted coding agent hosted-agents
"implement BDI", mental state modeling, belief-desire-intention bdi-mental-states
"start an LLM project", batch pipeline, task-model fit, project structure project-development

How to Load a Sub-Skill

  1. Identify the best-matching row above.
  2. Check if the local sub-skill SKILL.md exists:
    references/context-engineering-skills/skills/\x3Csub-skill>/SKILL.md
    
    • If the file exists locally: Read it directly.
    • If the file does NOT exist (e.g., submodule not initialized): Fetch from GitHub:
      https://raw.githubusercontent.com/muratcankoylan/Agent-Skills-for-Context-Engineering/main/skills/\x3Csub-skill>/SKILL.md
      
  3. Follow the instructions in that file. Most sub-skills have a references/ directory with detailed patterns — read those files only when the relevant pattern is needed (progressive disclosure).
  4. If the user's task spans two sub-skills (e.g., memory + multi-agent), read both SKILL.md files before responding.

Progressive Disclosure

Sub-skill SKILL.md files are compact routing documents. The deep content lives in each sub-skill's references/ folder. Load reference files only when the specific topic is active — do not bulk-load all references upfront.

Example:

references/context-engineering-skills/skills/context-optimization/references/

Read individual reference files from that directory only when the matching pattern is needed.

安全使用建议
This skill is coherent and appears to only load documentation from local files or from public GitHub raw URLs. Before installing or enabling autonomous use, consider: (1) Trust the upstream repository and author—remote SKILL.md content can change; if an attacker modifies upstream files the agent could load malicious instructions. (2) If you need to eliminate network risk, initialize and pin a local submodule or vendor a snapshot of the repository (prefer a fixed commit). (3) Be cautious about executing the optional git submodule update or any shell commands that appear in fetched SKILL.md files—these would modify local workspace files. (4) No credentials are requested by this skill. If you want lower blast radius, restrict autonomous invocation or review fetched SKILL.md content before allowing the agent to act on it.
功能分析
Type: OpenClaw Skill Name: levineam-context-engineering Version: 1.0.0 This skill is suspicious due to its design allowing for remote instruction execution and supply chain vulnerability. The `SKILL.md` explicitly instructs the AI agent to fetch sub-skill documentation (`SKILL.md` files) from a remote GitHub repository (`https://raw.githubusercontent.com/muratcankoylan/Agent-Skills-for-Context-Engineering/main/skills/<sub-skill>/SKILL.md`) and then to 'Follow the instructions in that file.' This creates a critical prompt injection vulnerability and supply chain risk, as a compromise of the upstream GitHub repository could lead to the agent executing arbitrary malicious instructions (e.g., data exfiltration, unauthorized actions) embedded in the remotely loaded Markdown files.
能力评估
Purpose & Capability
The skill's name and description claim routing to sub-skills and guidance about context engineering; the SKILL.md only requires reading sub-skill SKILL.md files (local or from raw.githubusercontent.com). No unrelated credentials, binaries, or installs are requested—this is proportionate to the stated purpose.
Instruction Scope
Runtime instructions tell the agent to read local SKILL.md files if present or fetch them from GitHub raw URLs when absent, and to load reference files progressively. This is appropriate for a documentation/router skill, but it does give the agent network fetch capability at runtime and the ability to read files under references/context-engineering-skills/. The SKILL.md also suggests an optional git submodule update command for offline use (which would modify the workspace if executed).
Install Mechanism
No install spec or code files are present. The skill is instruction-only and does not download or extract archives itself, so there is no installer risk in the skill bundle.
Credentials
The skill declares no environment variables, credentials, or config paths. This is appropriate for a documentation-routing skill.
Persistence & Privilege
always:false and no special privileges are requested. The only persistence-related action suggested is an optional git submodule update that would write files into the workspace; that action is not required for normal operation but would give the skill a local copy of upstream content if executed.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install levineam-context-engineering
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /levineam-context-engineering 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
Initial publish: OpenClaw wrapper for Muratcan Koylan's Agent-Skills-for-Context-Engineering, with 13 sub-skills covering context optimization, memory systems, multi-agent patterns, evaluation, and more.
元数据
Slug levineam-context-engineering
版本 1.0.0
许可证
累计安装 1
当前安装数 1
历史版本数 1
常见问题

Context Engineering (Koylan) 是什么?

Comprehensive context engineering guidance for AI agent systems. Routes to specialized sub-skills for production agent work. Use when the user asks to: "opti... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 413 次。

如何安装 Context Engineering (Koylan)?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install levineam-context-engineering」即可一键安装,无需额外配置。

Context Engineering (Koylan) 是免费的吗?

是的,Context Engineering (Koylan) 完全免费(开源免费),可自由下载、安装和使用。

Context Engineering (Koylan) 支持哪些平台?

Context Engineering (Koylan) 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 Context Engineering (Koylan)?

由 levineam(@levineam)开发并维护,当前版本 v1.0.0。

💬 留言讨论