persistent-memory

Capture observations, decisions, and errors across sessions with hybrid search.

Updated Mar 13, 2026

Search Tags:#context #session management #retrieval #memory #persistence #knowledge capture

Ready to run

npx skills add https://github.com/TECHKNOWMAD-LABS/cortex-research-suite --skill persistent-memory-techknowmad-labs

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: persistent-memory
Source: https://github.com/TECHKNOWMAD-LABS/cortex-research-suite/tree/main/skills/persistent-memory
Command: npx skills add https://github.com/TECHKNOWMAD-LABS/cortex-research-suite --skill persistent-memory-techknowmad-labs

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires sqlite3, json, hashlib, logging, datetime, pathlib, typing, enum, re, os, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill provides a robust, enterprise-grade memory system for AI agents, ensuring that crucial observations, decisions, and errors are captured and retrievable across sessions, preventing knowledge loss and improving continuity.

Core Features & Use Cases

Persistent Knowledge Capture: Stores observations, decisions, errors, and context across conversation boundaries.
Progressive Disclosure Retrieval: Efficiently retrieves information through layered search (index, timeline, full detail) to minimize token usage.
Hybrid Search: Combines full-text and semantic similarity for precise recall.
Use Case: An AI assistant working on a long-term project can recall specific architectural decisions made weeks ago, or retrieve the exact error message and its resolution from a previous debugging session, without needing to re-explain the entire context.

Quick Start

Use the persistent-memory skill to search your memory for "database migration errors".

Frequently Asked Questions about persistent-memory

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

FAQPage Schema

How do I retain context and session management for AI agents across multiple sessions?▼

Persistent memory captures observations, decisions, errors, and context across conversation boundaries using SQLite3, preventing knowledge loss in AI agents. It stores enterprise session data so assistants recall prior interactions without re-explaining context.

What is progressive disclosure retrieval and how does it minimize token usage?▼

Progressive disclosure retrieval layers search operations across index, timeline, and full detail to minimize token usage. This hybrid search approach combines full-text matching with semantic similarity scoring for precise, token-efficient recall of stored memory.

Can I use SQLite3 for enterprise persistent memory with automatic observation recording?▼

Yes, SQLite3 supports enterprise persistent memory storage with automatic observation recording. The system tracks dual sessions for cross-session continuity and includes AI-powered compression to manage storage efficiently.

How do I search captured knowledge and decisions from previous debugging sessions?▼

Hybrid search combines full-text and semantic similarity scoring to retrieve specific architectural decisions or error resolutions from previous sessions. You query stored memory directly to recall exact context without manual filtering.

Does persistent memory support privacy controls for sensitive enterprise context?▼

Yes, persistent memory includes privacy controls for managing sensitive enterprise context. The system ensures captured observations, decisions, and errors are stored with configurable access boundaries across session boundaries.

When should I not use a persistent memory system for AI session management?▼

Avoid persistent memory when conversations are strictly stateless or when storage dependencies like SQLite3 are unavailable. If sessions require no cross-conversation continuity or knowledge capture, the overhead of hybrid search offers limited value.