Store and retrieve persistent memories across sessions using Markdown files with YAML front matter.
npx skills add https://github.com/wszqkzqk/openmemory --skill openmemory-wszqkzqk
Please help me install this Agent Skill. Skill: openmemory Source: https://github.com/wszqkzqk/openmemory/tree/main/skills/openmemory Command: npx skills add https://github.com/wszqkzqk/openmemory --skill openmemory-wszqkzqk
OpenMemory provides a filesystem-based persistent memory layer for coding agents, enabling context, decisions, and preferences to survive session boundaries.
Store a memory entry at the start of a task and retrieve it later to maintain continuity.
High-intent search queries and answers about installing and using this skill.
You can record architectural decisions and task state by storing memory entries as Markdown files with YAML front matter, then retrieve them later using memory_list, memory_get, and memory_search tools to maintain project continuity.
You can record architectural decisions and task state by storing memory entries as Markdown files with YAML front matter, then retrieve them later using memory_list, memory_get, and memory_search tools to maintain project continuity.
This approach uses Markdown files with YAML front matter as a filesystem-based persistent memory layer, applying to project and global scopes to maintain context and preferences without requiring a database.
Yes, it supports cross-session and cross-project memory that can be browsed and updated using memory_list, memory_search, and memory_store, allowing you to retrieve context across different project scopes.
Limitations include following guardrails to avoid storing secrets and maintaining a clear paper trail when conflicting memories arise across the filesystem-based Markdown memory layer.
forwardimpactBuild and serve static documentation sites from Markdown with YAML front matter.
freedomdpStore and retrieve persistent, searchable memories via an MCP server.
qdoubleCoordinate multi-agent software development workflows through structured plan files.
muratcankoylanOffload oversized context data to filesystem files with grep/read_file retrieval.
