memory

Stores and recalls per-role knowledge via markdown files and daily logs.

18|9|Updated Apr 8, 2026
One-click install
npx skills add https://github.com/arozumenko/sdlc-skills --skill memory-arozumenko
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: memory
Source: https://github.com/arozumenko/sdlc-skills/tree/main/skills/memory
Command: npx skills add https://github.com/arozumenko/sdlc-skills --skill memory-arozumenko

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It prevents important decisions, preferences, and project context from getting lost between sessions by giving agents a durable place to store and recall knowledge.

Core Features & Use Cases

  • Per-role persistent memory: Store durable, agent-specific facts and preferences as plain markdown under a role-scoped memory directory.
  • Daily logging: Append one-line-a-day learnings to keep a lightweight audit trail of what changed recently.
  • Curated index with types: Maintain an index of curated entries (user, feedback, project, reference) and read only what’s needed to keep context bounded.

Quick Start

Use the memory skill whenever the user says remember this, log this, asks what you learned yesterday, or you identify something worth keeping across sessions.

Frequently Asked Questions about memory

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

FAQPage Schema
How do I give an AI agent persistent memory across IDE sessions?

Persistent memory is maintained by storing per-role durable knowledge as curated markdown entries and append-only daily logs under a role-scoped memory directory. This ensures decisions and project context survive across sessions.

How do I keep an audit trail of daily learnings for an agent workflow?

An audit trail of daily learnings is kept by appending one-line entries to daily logs. This creates a lightweight record of what changed recently without bloating the context window.

Can I store project context and decisions as markdown for an agent to read later?

Yes, project context and decisions are stored as plain markdown by updating a curated index file. The index categorizes entries by type so agents can read only what is needed.

What is the best way to maintain bounded context when recalling agent memory?

Bounded context is maintained by reading only the curated index entries and the most recent daily logs. This prevents injecting the entire memory history into the agent's context.

Does this approach to agent memory work without external dependencies?

Yes, this memory approach works without external dependencies by relying entirely on the local file system. It writes and reads markdown files directly within the project directory.

When do I need to use a curated knowledge index for agent memory?

A curated knowledge index is needed when an agent must distinguish between user preferences, feedback, project briefings, and references. It allows the agent to selectively load specific entry types to keep context bounded.