agent-context-system

Automate persistent project context for AI coding agents using markdown files.

23|3|Updated Feb 7, 2026
One-click install
npx skills add https://github.com/AndreaGriffiths11/agent-context-system --skill agent-context-system
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-context-system
Source: https://github.com/AndreaGriffiths11/agent-context-system/tree/main
Command: npx skills add https://github.com/AndreaGriffiths11/agent-context-system --skill agent-context-system

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

AI coding agents forget context between sessions; this Skill provides a persistent, shareable project memory.

Core Features & Use Cases

  • Two-file memory model: AGENTS.md (committed, shared) + .agents.local.md (gitignored, personal)
  • Cross-agent consistency: works across Claude Code, Cursor, Copilot, Windsurf
  • Session logging and promotion: reads at session start and promotes stable patterns over time
  • No plugins or external infra: plain markdown in the repo

Quick Start

Initialize the agent-context system by running the init script to create AGENTS.md and the personal scratchpad, then begin sessions.

Frequently Asked Questions about agent-context-system

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I keep persistent project context for AI coding agents across sessions?

You can maintain persistent project context for AI coding agents by using a two-file markdown system: AGENTS.md for committed shared memory and .agents.local.md for personal gitignored notes, read at session start to preserve continuity.

Does this context management approach work with Claude Code, Cursor, Copilot, and Windsurf?

Yes, this agent context system applies cross-agent consistency across Claude Code, Cursor, Copilot, and Windsurf, maintaining continuity without requiring plugins or external infrastructure.

How do I set up persistent memory for AI agents in my repository?

Run the init script to create the AGENTS.md shared memory file and the .agents.local.md personal scratchpad, then begin your coding sessions to start logging context automatically.

What's the best way to share AI agent context with my team without external infrastructure?

Commit the AGENTS.md file to your repository for shared team context while keeping .agents.local.md gitignored for personal scratchpad notes, using plain markdown without plugins or infrastructure.

How does session logging and pattern promotion work for agent memory?

Session logging reads project context at session start and promotes stable patterns over time, consolidating repeated context into durable memory entries within the two-file markdown system.

Do I need any plugins or external infrastructure to maintain agent context?

No plugins or external infrastructure are required; the system uses plain markdown files stored directly in your repository to maintain persistent context across AI agent sessions.