dory-memory

Name: dory-memory
Availability: InStock
Author: wewetv1987-cell

Record exact user instructions in file-based workspace state across session resets.

Updated Feb 16, 2026

Search Tags:#state management #memory #session continuity #task tracking #workspace setup #agent continuity #context resets

Ready to run

npx skills add https://github.com/wewetv1987-cell/openclaw-backup --skill dory-memory

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: dory-memory
Source: https://github.com/wewetv1987-cell/openclaw-backup/tree/main/skills/dory-memory
Command: npx skills add https://github.com/wewetv1987-cell/openclaw-backup --skill dory-memory

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) and assets (resource) components.

What problem does it solve?

This Skill prevents AI agents from losing task context between sessions by moving memory from conversation history into durable files on disk.

Core Features & Use Cases

Exact-word task capture: Record the user's instructions verbatim before interpreting them.
Boot-time state recovery: Check HOLD, ACTIVE, DECISIONS, and recent work files to resume safely after resets.
Workspace memory structure: Set up standardized files for identity, long-term memory, lessons, staging, and operational indexing.
Use Case: A support agent can restart mid-project, read the workspace files, and continue without repeating questions or missing blocked items.

Quick Start

Tell the agent to initialize the Dory-proof workspace and record the user's exact request in state/ACTIVE.md before doing anything else.

Frequently Asked Questions about dory-memory

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

FAQPage Schema

How do I preserve AI agent context and task tracking across session resets?▼

To preserve agent context across session resets, move memory from conversation history into durable file-based workspace state with standardized files for active tasks, decisions, and long-term memory.

What is file-based workspace state for AI agents?▼

File-based workspace state is a root-level memory system using ACTIVE, HOLD, DECISIONS, and MEMORY files to capture exact user instructions and enable boot-time state recovery for continuous task tracking.

How do I set up agent continuity and memory curation for long-term projects?▼

Set up agent continuity by initializing a standardized workspace structure that records exact-word user instructions, manages blocked items, logs decisions, and curates long-term memory across operational index files.

Can I use file-based state management to resume a blocked support agent mid-project?▼

Yes, a support agent can restart mid-project by reading workspace state files at boot, checking HOLD and ACTIVE items, and continuing safely without repeating questions or missing blocked tasks.

Does agent state management work without relying on conversational context?▼

Yes, this state management approach works without conversational context by recording exact user instructions verbatim in disk files, enforcing boot sequencing, and applying conflict-priority rules to maintain continuity.

What are the limitations of using file-based memory for session continuity?▼

File-based memory requires a root-level state system with specific file templates and strict exact-word capture rules, meaning agents must follow boot sequencing and conflict-priority protocols precisely to avoid context loss.