Persist user preferences and notes in Markdown prompt-include memory files.
npx skills add https://github.com/agent0ai/space-agent --skill memory-agent0ai
Please help me install this Agent Skill. Skill: Memory Source: https://github.com/agent0ai/space-agent/tree/main/app/L0/_all/mod/_core/memory/ext/skills/memory Command: npx skills add https://github.com/agent0ai/space-agent --skill memory-agent0ai
This Skill helps preserve important user-specific behavior and evolving notes so the agent can act consistently across future conversations instead of relearning preferences and context each time.
Tell the agent what durable preference, rule, or project detail to retain and ask it to update the appropriate ~/memory include file accordingly.
High-intent search queries and answers about installing and using this skill.
To persist agent behavior across sessions, store durable preferences in user-scoped Markdown prompt-include files that load into the system prompt automatically. This prevents preference loss and redundant re-ask patterns in ongoing conversations.
Tracking evolving project context requires writing rolling notes into transient Markdown include files. These files retain short- to medium-lived facts and reminders, keeping frequently updated information current without bloating durable system instructions.
Prompt-include memory separates system rules from temporary facts by routing them into distinct file patterns: durable behaviors go into behavior.system.include.md, while rolling notes go into memories.transient.include.md or similar transient files.
No, you do not need YAML frontmatter in Markdown files for context management. Memory include files must contain plain Markdown include bodies without YAML frontmatter to ensure correct runtime loading into the system prompt.
You should update memory files when user-scoped behaviors, project context, or reminders change. Following memory hygiene rules, incrementally mutate the files to keep content concise, specific, and current for consistent agent behavior.
An assistant loses preferences in new conversations because session data resets. Storing durable instructions and evolving notes in user-scoped Markdown prompt-include files prevents this context loss and redundant relearning across sessions.
binary16labsPersist user preferences and notes in prompt-include memory markdown files.
GoBeromsuStructure and persist project-wide context rules across agent sessions.
guanyangFormat Markdown files with YAML frontmatter and normalized typography.
bugrabilgeScan project files and store context, patterns, and decisions in markdown memory files.
