skills.rest
p-diogo
p-diogoCOMMUNITY
Explore all skills by this author →
GitHub

totalreclaw

Store and retrieve encrypted long-term memory for AI agents.

1|2|Updated Feb 23, 2026
Search Tags:#knowledge graph#agent memory#openclaw#e2ee#encrypted recall#portable export
One-click install
npx skills add https://github.com/p-diogo/totalreclaw --skill totalreclaw
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: totalreclaw
Source: https://github.com/p-diogo/totalreclaw/tree/main/skill
Command: npx skills add https://github.com/p-diogo/totalreclaw --skill totalreclaw

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

TotalReclaw enables AI agents to remember preferences, decisions, and facts across conversations while ensuring the server can never read plaintext.

Core Features & Use Cases

  • End-to-end encrypted memory (E2EE): Encrypts memories client-side with XChaCha20-Poly1305 so the relay/server cannot decrypt them.
  • Structured memory taxonomy & provenance: Stores memory entries with types (e.g., claim, preference, directive, commitment, episode, summary) plus provenance/scope/volatility to improve recall quality.
  • Graph-anchored retrieval + portable export: Uses an indexed decentralized storage/search setup and supports exporting memories for portability and migration.

Quick Start

Ask your AI assistant: "Install the totalreclaw skill from ClawHub and set up my account, then remember that I prefer dark mode and recall it next time."

Frequently Asked Questions about totalreclaw

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

FAQPage Schema
How do I give my AI agent end-to-end encrypted memory across conversations?▼

End-to-end encrypted memory for AI agents is provided by encrypting facts client-side using XChaCha20-Poly1305, ensuring the server cannot access plaintext while storing preferences and decisions across sessions.

How does encrypted recall work for agent memory without server-side plaintext access?▼

Encrypted recall works by using indexed decentralized storage and search to retrieve client-side encrypted memories, preventing the relay or server from decrypting stored user-specific facts and preferences.

How do I export and migrate agent memory with a structured knowledge graph?▼

Portable export of agent memory is supported through a structured taxonomy that stores memory entries with types, provenance, scope, and volatility, allowing you to migrate encrypted knowledge graphs across environments.

Can I use OpenClaw skills with hybrid CLI workflows for managing agent memory?▼

OpenClaw skill usage is supported alongside hybrid CLI workflows for installing, pairing, remembering, recalling, forgetting, consolidating, and checking quota of encrypted agent memories across conversations.

What are the limitations of using client-side encrypted storage for agent memory?▼

Limitations include the strict requirement to avoid cleartext memory file writes, meaning all operations must enforce phrase safety and rely on encrypted client-side storage to maintain zero-knowledge server access.

Do I need YAML-frontmatter skill metadata to set up encrypted agent memory?▼

YAML-frontmatter-discoverable skill metadata is required alongside operational instructions and tool mappings to correctly configure encrypted long-term memory and enforce safe phrase usage for AI agents.

Related Skills

Explore More Software Engineering Skills→
XCmiaowXCmiaow

template-skill

Create Claude Code skills with YAML frontmatter and instruction stubs.

Community
Basic
hongmaple0820hongmaple0820
9

networkx

Create, analyze, and visualize graphs with Python NetworkX.

Community
Advanced
shreed27shreed27

memory

Store, recall, and search persistent user data with a TypeScript API.

Community
Advanced
onichandameonichandame

client-side-encryption

Encrypt user data client-side with AES-256-GCM and PBKDF2 in SvelteKit.

Community
Advanced