cortex

Trace file changes and commits to generate Cortex session memory files.

2|Updated Oct 26, 2025
One-click install
npx skills add https://github.com/bacoco/EvolveSkill --skill cortex
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cortex
Source: https://github.com/bacoco/EvolveSkill/tree/main/.claude/skills/cortex
Command: npx skills add https://github.com/bacoco/EvolveSkill --skill cortex

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Claude often loses context across sessions or needs to be reminded of past work, leading to repetitive explanations and wasted time. Cortex eliminates this by providing an automatic, persistent memory system.

Core Features & Use Cases

  • Automatic Session Tracing: Tracks all file changes, Git operations, and major events in your project.
  • Seamless Agent Handoffs: Creates a .cortex_handoff.md file, enabling Claude to instantly pick up where it left off in any new session.
  • Inter-Skill Communication: Provides a Python API for different skills to share data and patterns, fostering a more integrated AI workflow.
  • Use Case: Imagine you're working on a complex feature over several days. Cortex automatically logs all your code changes, commits, and decisions. When you return, Claude instantly knows exactly where you left off and what to do next, without you having to re-explain.

Quick Start

Start a new coding session in a Git repository. Cortex will automatically track your changes and provide a .cortex_handoff.md file, summarizing your last session and next steps.

Frequently Asked Questions about cortex

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

FAQPage Schema
How do I automatically track file changes and git commits across coding sessions?

Cortex automatically traces file modifications, git operations, and session events within your repository, generating logs that capture your complete project history. It eliminates manual context-tracking by recording all changes in machine-readable formats for instant retrieval.

Can Claude pick up where I left off in a new session without re-explaining my work?

Yes. Cortex generates a `.cortex_handoff.md` file that summarizes your last session, decisions, and next steps. Claude reads this handoff automatically, enabling seamless continuity without repetitive explanations across sessions.

How does Cortex enable different AI skills to share data and patterns?

Cortex provides a Python API for inter-skill communication, allowing different skills to exchange session data, patterns, and insights. This integration fosters coordinated AI workflows where multiple skills build on shared context and memory.

What do I need to set up before using Cortex in my git repository?

Cortex requires git hooks and a Python-based API within your repository. It operates directly on git repositories with no external dependencies, automatically activating when you start a coding session.

Does Cortex work with other LLMs beyond Claude?

Cortex is designed for cross-LLM compatibility. Its memory formats—`.cortex_log.md`, `.cortex_status.json`, and `.cortex_handoff.md`—are standards-based and agnostic, enabling any AI system to leverage the captured context.

What happens if I need to analyze patterns from my session memory to generate new skills?

Cortex triggers pattern analysis via Synapse, which examines your session logs to identify workflows and generate memory-driven skills. This enables automated skill creation based on detected usage patterns.