harness-design-patternsAgent harness 架构设计模式知识库。当设计多阶段任务的上下文传递、选择多 agent 协调方式、或规划 hook 系统架构时使用。纯设计指南，无可执行代码。不用于安装 hook 脚本（用 agent-hooks）或运维监控（用 agent-ops）。
Install via ClawdBot CLI:
clawdbot install lanyasheng/harness-design-patternsGrade Limited — based on market validation, documentation quality, package completeness, maintenance status, and authenticity signals.
Calls external URL not in known-safe list
https://www.anthropic.com/engineering/effective-harnesses-for-long-running-agentAudited Apr 17, 2026 · audit v1.0
Generated Apr 24, 2026
An AI agent needs to execute a plan-implement-verify pipeline across multiple turns. Use Handoff documents to pass context between stages and Compaction extraction to preserve knowledge before memory compression.
Deciding whether to use a coordinator agent, forked sub-agents, or a swarm for parallel tasks. The three delegation patterns guide selection based on task dependency and communication overhead.
Multiple hooks cause interference and performance issues. Use Component-Scoped Hooks to restrict hooks to specific tasks and Hook Runtime Profiles to adjust hook intensity per environment.
Agent forgets important context after memory compaction or between sessions. Apply Handoff documents for explicit context handover and Compaction extraction to retrieve critical info before compression.
Simple tasks incur high overhead from complex agent loops. Use Adaptive Complexity to triage tasks and automatically select minimal execution mode, reducing latency and cost.
Offer harness architecture design consulting to enterprises building multi-agent systems. Provide expertise in pattern selection and implementation guidance.
Develop training programs on harness design patterns for AI engineers. Sell access to course materials, workshops, and certification exams.
Package harness design patterns into a SaaS tool that analyzes agent architectures and recommends patterns. Charge subscription for automated design audits.
💬 Integration Tip
Combine with agent-hooks skill for executable hook scripts and agent-ops for operational monitoring to create a complete harness lifecycle solution.
Scored Apr 19, 2026
Control desktop applications on Windows — launch, close, focus, resize, move windows, simulate keyboard/mouse input, manage processes, control VSCode, read clipboard, and capture screen info. Use when the user wants to interact with any running program, switch windows, type text, press shortcuts, open files in VSCode, manage running processes, or get system display information.
Conduct rigorous, adversarial code reviews with zero tolerance for mediocrity. Use when users ask to "critically review" my code or a PR, "critique my code", "find issues in my code", or "what's wrong with this code". Identifies security holes, lazy patterns, edge case failures, and bad practices across Python, R, JavaScript/TypeScript, SQL, and front-end code. Scrutinizes error handling, type safety, performance, accessibility, and code quality. Provides structured feedback with severity tiers (Blocking, Required, Suggestions) and specific, actionable recommendations.
Coding style memory that adapts to your preferences, conventions, and patterns for consistent coding.
Pragmatic coding standards for writing clean, maintainable code — naming, functions, structure, anti-patterns, and pre-edit safety checks. Use when writing new code, refactoring existing code, reviewing code quality, or establishing coding standards.
Claude Code integration for OpenClaw. This skill provides interfaces to: - Query Claude Code documentation from https://code.claude.com/docs - Manage subagents and coding tasks - Execute AI-assisted coding workflows - Access best practices and common workflows Use this skill when users want to: - Get help with coding tasks - Query Claude Code documentation - Manage AI-assisted development workflows - Execute complex programming tasks
Plan, draft, version, and refine written content with enforced versioning and quality audits.