跳到正文
TakoAPI
返回技能列表

agent-context-system

由 andreagriffiths11 提供

Coding Agents & IDEs
566下载量由 clawskills.sh 提供
0浏览量
0 个赞登录后即可为这个技能点赞
—暂无评分
你的评分

简介

A persistent local-only memory system for AI coding agents.

描述

该技能的功能

A two-file convention for giving AI coding agents persistent memory across sessions. AGENTS.md holds committed project knowledge shared with the team; .agents.local.md is a gitignored personal scratchpad that grows session by session. No plugins, servers, or background processes required.

Unlike tool-specific memory features, the two-file convention works across Claude Code, Cursor, Copilot, and Windsurf in the same project without any additional tooling.

何时使用

  • Preserving project context so agents don't cold-start each session
  • Sharing architecture patterns and gotchas with teammates via git
  • Logging dead ends so agents don't retry approaches that already failed
  • Coordinating parallel subagents that all share the same project knowledge
  • Gradually promoting recurring session learnings into permanent project docs

示例工作流

输入

User starts a new coding session on an existing project

智能体
  1. 1

    Agent reads AGENTS.md for committed project patterns, commands, and boundaries

  2. 2

    Agent reads .agents.local.md for personal preferences, past dead ends, and session logs

  3. 3

    Agent completes the task using accumulated context from both files

  4. 4

    Agent proposes a session log entry to the user before appending to .agents.local.md

  5. 5

    Agent writes the approved entry and flags any patterns that have recurred across 3+ sessions as ready to promote

输出

Session knowledge is logged locally; stable patterns are flagged for the user to promote into AGENTS.md

安装

clawhub install andreagriffiths11/agent-context-system

运行 CLI 命令,或把这段提示词粘贴到 Claude Code、Cursor、Codex 或你在用的任意编程智能体里。

链接