# Project Memory Practices

> Guidelines for using Claude app memory (including project-scoped memory and incognito chat) to maintain context safely and intentionally across ongoing work.

- Skill: `uygnoey/project-memory-practices` (Agent Skill)
- Install (CLI): `npx skillmds@latest add uygnoey/project-memory-practices`
- Raw SKILL.md: https://api.skillmd.com/api/skills/uygnoey/project-memory-practices/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: uygnoey (https://skillmd.com/u/uygnoey)
- Updated: 2026-09-21
- Page: https://skillmd.com/skills/uygnoey/project-memory-practices

---


# project-memory-practices

## Instructions

- Apply the post’s guidance as product usage practices (not API behavior): treat memory as optional and user-controlled, and keep sensitive work scoped appropriately.    - Prefer project-scoped context: if you are working in multiple initiatives, keep each initiative in its own project so remembered context does not mix.    - Use incognito chats when you do not want a conversation to appear in history or add to memory.    - When memory is enabled, periodically review and edit the memory summary, and tell Claude what to focus on or ignore.

## Examples

- "We have two parallel initiatives. Summarize what you remember for Project A only, and do not use context from other projects."    - "This is a one-off confidential brainstorm—please treat this as an incognito-style conversation and don’t carry details forward."    - "Here is what I want you to stop remembering about our process: …"
## Source

- https://claude.com/blog/memory