skills.rest
yishan-io
yishan-ioOFFICIAL
Explore all skills by this author →
GitHub

context-memory

Search and update durable memory files in the .my-context directory.

7|2|Updated Apr 23, 2026
Search Tags:#architecture#documentation#context#knowledge management#memory#decision tracking
One-click install
npx skills add https://github.com/yishan-io/yishan-mono --skill context-memory-yishan-io
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: context-memory
Source: https://github.com/yishan-io/yishan-mono/tree/main/packages/pi-memory/skills/context-memory
Command: npx skills add https://github.com/yishan-io/yishan-mono --skill context-memory-yishan-io

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill solves the problem of fragmented project knowledge by providing a structured, durable repository for decisions, architecture notes, and non-obvious facts that must persist across multiple tasks and sessions.

Core Features & Use Cases

  • Durable Knowledge Storage: Maintains a centralized memory layer in .my-context/MEMORY.md for long-term project context.
  • Decision Tracking: Records locked decisions and root causes to prevent the re-litigation of past architectural choices.
  • Use Case: When a developer discovers a recurring environment trap or a critical architectural invariant, they use this skill to document it so that future agents or team members can avoid the same pitfalls without needing to rediscover the information.

Quick Start

Use the context-memory skill to search for existing architectural decisions regarding the authentication module before starting your new task.

Frequently Asked Questions about context-memory

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

FAQPage Schema
How do I maintain persistent project memory for AI agents across different sessions?▼

Maintain persistent project memory by searching and updating durable memory files within a project directory to preserve architectural decisions, root causes, and recurring workflow traps across independent workstreams and sessions.

What is the best way to track architectural decisions so AI agents don't re-litigate past choices?▼

Track architectural decisions by recording locked decisions and root causes in a centralized memory layer, preventing autonomous agents from re-litigating past architectural choices during future independent workstreams.

How do I document recurring environment traps to prevent autonomous agents from rediscovering them?▼

Document recurring environment traps by recording non-obvious facts and critical invariants in a durable knowledge repository, ensuring future agents avoid the same pitfalls without needing to rediscover the information independently.

Do I need project-specific memory documentation to use a context memory system?▼

Project-specific memory documentation is required to ensure consistent context availability for autonomous agents, enabling the durable memory system to search and update knowledge accurately across independent workstreams.

When should I search existing architectural decisions before starting a new development task?▼

Search existing architectural decisions before starting a new development task to retrieve prior context from durable memory files, preventing the re-litigation of locked decisions and avoiding previously documented workflow traps.

Related Skills

Explore More Software Engineering Skills→
erclxerclx
2

claude-memory-capture

Extract session insights into structured Markdown memory files.

Community
Intermediate
insightfloinsightflo
17

discuss

Collect user decisions to generate structured CONTEXT.md files for downstream agents.

Community
Intermediate
phrazzldphrazzld
13

distill

Distill repository knowledge into a concise CLAUDE.md onboarding guide.

Community
Intermediate
PlaidmustachePlaidmustache

qmd

Search local knowledge bases with hybrid BM25 and vector retrieval.

Community
Advanced