skills.rest
opensquilla
opensquillaOFFICIAL
Explore all skills by this author →
WebsiteGitHub

memory

Manage durable recall and updates of OpenSquilla memory via USER.md and MEMORY.md files.

6.6k|511|Updated May 6, 2026
Search Tags:#recall#knowledge base#memory#user profile#long-term preferences#durable notes
One-click install
npx skills add https://github.com/opensquilla/opensquilla --skill memory-opensquilla
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: memory
Source: https://github.com/opensquilla/opensquilla/tree/main/src/opensquilla/skills/bundled/memory
Command: npx skills add https://github.com/opensquilla/opensquilla --skill memory-opensquilla

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The memory skill helps you reliably store and retrieve durable facts so the agent can recall preferences, profile details, and long-term notes across sessions instead of asking the same questions again.

Core Features & Use Cases

  • Recall durable memory: Use injected USER.md for current identity/profile questions and search curated memory files for historical non-profile recall.
  • Remember, update, and correct: Update user profile fields via USER.md, create daily or named notes under memory/YYYY-MM-DD.md or memory/**/*.md, and update long-term curated facts in MEMORY.md by rewriting the full content.
  • Forget safely: Find the relevant source first, then delete an entire memory source file when supported or edit the exact lines/paths when write/delete tools are not available.
  • Bounded, safe storage: Avoid saving deliverables, secrets, tokens, private keys, or full credential contents; confirm updates only after the write or delete succeeds.

Quick Start

Ask: "Remember that my timezone is America/New_York and update my profile, then recall it when I ask about my timezone."

Frequently Asked Questions about memory

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

FAQPage Schema
How does an agent recall long-term preferences and user profile details across sessions?▼

To remember durable facts across turns, you ask the agent to update your profile in USER.md, append notes to dated files, or rewrite curated facts in MEMORY.md, confirming the action only after the write succeeds.

How do I update or correct a long-term memory fact in my knowledge base?▼

To update or correct a long-term memory fact, you rewrite the full content of the relevant entry in MEMORY.md or edit the exact lines in the specific source file, ensuring the update is confirmed after the write operation succeeds.

Can I safely delete a memory file without removing sensitive credentials or deliverables?▼

You can safely forget memory by deleting an entire source file or editing exact lines, but the system avoids saving deliverables, secrets, tokens, or private keys, bounding storage to non-sensitive durable facts only.

What is the best way to store durable notes so an agent remembers them across conversations?▼

The best way to store durable notes is creating daily or named notes under memory/YYYY-MM-DD.md or memory/**/*.md paths, which allows the agent to search and retrieve these scoped facts in future interactions.

When should I not use agent memory for storing information?▼

You should not use agent memory for storing deliverables, secrets, tokens, private keys, or full credential contents, as the system boundaries explicitly exclude sensitive data to maintain safe and bounded profile and preference storage.

Related Skills

Explore More Personal & Entrepreneur Skills→
adamrdrewadamrdrew
1

create-memory

Create memory.md with a standard header and entries section when missing.

Community
Basic
tocadaoncamodtocadaoncamod

memoria-contexto

Store and prioritize non-sensitive user attributes across conversational sessions.

Community
Intermediate
bcheng004bcheng004

agent-memory

Persist agent conversation checkpoints and user preferences via Databricks Lakebase.

Community
Intermediate
hyperb1isshyperb1iss
47

sibyl

Store coding patterns, learnings, and tasks in a knowledge graph.

Community
Advanced