eh-context-memory上下文窗口管理与跨 session 知识传递。当需要跨阶段传递决策、压缩前抢救知识时使用。不适用于工具重试（用 tool-governance）或多 agent 协调（用 multi-agent）。参见 execution-loop（阶段边界）。
Install via ClawdBot CLI:
clawdbot install lanyasheng/eh-context-memoryGrade Fair — based on market validation, documentation quality, package completeness, maintenance status, and authenticity signals.
Generated Apr 24, 2026
A multi-turn customer support session where the AI agent passes context (e.g., issue details, customer sentiment) to the next agent or human via handoff documents, ensuring seamless continuation without repeating information.
An AI writing assistant generates a multi-chapter report; before each token budget compaction, it extracts key decisions and data into a compaction file, preserving narrative coherence across sessions.
A diagnostic AI processes patient history across multiple stages (symptoms, lab results, specialist input); token budget estimation prevents context overflow, and handoff documents pass critical findings between stages.
A legal AI reviews large contract sets; it uses filesystem as working memory to store partial analyses, and compaction extraction saves key clauses before context reset, enabling consistent multi-session review.
An AI coding assistant helps debug and refactor a large codebase across multiple sessions; it saves unresolved issues and decisions via handoff documents, and monitors context to avoid losing track of the file structure.
Offer context management as a premium feature in an AI agent platform, charging per user or per token/compaction operation. Recurring revenue from subscriptions and upsells for higher token budgets.
License the context-memory skill package to enterprises with custom workflows (e.g., legal, healthcare). Provide integration support and training. Revenue from licensing fees and professional services.
List the skill on an AI agent marketplace (e.g., LangChain Hub) and take a 20-30% commission on each sale. Volume-driven revenue with low customer acquisition cost.
💬 Integration Tip
Leverage the provided shell scripts (context-usage.sh and compaction-extract.sh) as building blocks; start with Handoff documents pattern for simple cross-session transfer, then layer on token budgeting.
Scored Jul 6, 2026
Meta-skill for AI agent self-improvement. Analyzes runtime logs to detect error patterns, regressions, and inefficiencies, then generates structured improvem...
Stop waiting for prompts. Keep working.
Turn OpenClaw into a learning-loop agent with seeded workspace rules, skill promotion, reflective memory, and proactive maintenance.
Local Python orchestration skill: multi-agent workflows via shared blackboard file, permission gating, token budget scripts, and persistent project context....
Meta-agent skill for orchestrating complex tasks through autonomous sub-agents. Decomposes macro tasks into subtasks, spawns specialized sub-agents with dynamically generated SKILL.md files, coordinates file-based communication, consolidates results, and dissolves agents upon completion. MANDATORY TRIGGERS: orchestrate, multi-agent, decompose task, spawn agents, sub-agents, parallel agents, agent coordination, task breakdown, meta-agent, agent factory, delegate tasks
Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.