clawtext

Provide durable memory and context protection for OpenClaw agents.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/PsiClawOps/ClawText --skill clawtext
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: clawtext
Source: https://github.com/PsiClawOps/ClawText/tree/main
Command: npx skills add https://github.com/PsiClawOps/ClawText --skill clawtext

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires openclaw, js-yaml, and includes scripts (resource) and references (resource) components.

What problem does it solve?

ClawText solves the challenge of preserving important context across sessions for OpenClaw agents by providing durable memory, retrieval, ingest, and continuity handoffs that keep what matters and minimize what doesn't.

Core Features & Use Cases

  • Working memory: automatic prompt-time memory injection and retrieval.
  • Ingest and deduplicate external sources to build durable memory.
  • Operational learning: capture failures, patterns, and transfer knowledge.
  • Continuity transfer: structured handoffs to move active work across threads/sessions without context loss.
  • Strong data hygiene: protect secrets and enforce auditability in memory episodes.

Quick Start

Install the ClawText plugin in OpenClaw, enable the RAG skill, restart gateway, and begin ingesting memory.

Frequently Asked Questions about clawtext

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

FAQPage Schema
How do I preserve agent context across sessions?

Preserving agent context across sessions requires durable memory and structured continuity handoffs that capture active work and transfer it without loss. This approach uses frontmatter-based contracts to move context across threads while minimizing irrelevant data.

What is proactive context protection for AI agents?

Proactive context protection is a mechanism that safeguards important memory during ingestion and recall. It enforces data hygiene by protecting secrets and ensuring auditability within stored memory episodes.

How do I ingest and deduplicate external sources for agent memory?

Ingesting and deduplicating external sources for agent memory involves applying a retrieval workflow that filters redundant data during capture. This builds a durable memory store by keeping operational knowledge and patterns while discarding duplicates.

Does OpenClaw support durable working memory injection?

OpenClaw supports durable working memory through automatic prompt-time memory injection and retrieval. This enables agents to access operational learning and captured failures dynamically during active sessions.

What's the best way to build a RAG memory system for autonomous agents?

Building a RAG memory system for autonomous agents involves combining durable memory capture with proactive context protection. This method ensures continuity transfer and data hygiene while ingesting and deduplicating external sources.

Why does my agent lose operational knowledge between threads?

Agents lose operational knowledge between threads due to missing continuity handoffs that transfer active work across sessions. Without structured memory capture, failures and patterns are not ingested into a durable store for later recall.