steering-memory
OfficialEnforce memory rules to shape AI behavior.
Software Engineering#safety#memory#session-management#learnings#steering#intent-recognition#domain-mapping
Authoragentflock
Version1.0.0
Installs0
System Documentation
What problem does it solve?
Myna's behavior can drift without a structured memory model. This Skill defines a three-layer memory system with explicit precedence, session-start loading, and domain-aware rules to keep outputs consistent, safe, and aligned with user preferences.
Core Features & Use Cases
- Three-Layer Precedence: Hard rules, user bootstrap data from CLAUDE.md, and emergent learnings from _meta/learnings domains.
- Session-Start Loading: Load applicable learnings at the start of every session to set context.
- Domain Mapping & Intent: Resolve /myna:learn intents using domain-specific files and apply output boundaries and refusal for factual entries when needed.
Quick Start
Activate the memory model in a new session and allow it to load learnings and enforce the rules.
Dependency Matrix
Required Modules
None requiredComponents
Standard package💻 Claude Code Installation
Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.
Please help me install this Skill: Name: steering-memory Download link: https://github.com/agentflock/myna/archive/main.zip#steering-memory Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
Agent Skills Search Helper
Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.