What happens if my project's MEMORY.md file gets too long?

The skill tracks file length and issues a warning when you reach 180 lines, encouraging a review to keep the context window efficient.

Can I use this skill to store API keys or secrets?

No. You should never use /si:remember for sensitive data like credentials or tokens, as these are stored in plain text files.

When should I promote a memory to a rule?

If a memory represents a strict convention or an 'always/never' directive, promoting it to CLAUDE.md ensures it is enforced with higher priority than standard memory.

How does /si:remember differ from standard auto-memory?

While Claude captures some context automatically, this skill allows for explicit, manual anchoring of specific insights that are too important to rely on Claude's default observation.

Where is the saved knowledge actually stored?

Entries are saved to a MEMORY.md file located within the project's local .claude directory, ensuring context stays isolated to the relevant project.

Persistent Knowledge Memory

Name: Persistent Knowledge Memory
Author: zhangzhang-111-i

byzhangzhang-111-i

0•

Productivity & Workflow

Explicitly saves critical project insights and discoveries to Claude's long-term memory to ensure important context is never lost.

The Persistent Knowledge Memory skill provides a structured mechanism for developers to anchor hard-won insights, project-specific conventions, and debugging breakthroughs directly into Claude's auto-memory system. By utilizing the /si:remember command, users can bypass the uncertainty of automatic context capture, ensuring that critical architectural decisions or tool-specific 'gotchas' remain accessible across future sessions. The skill intelligently manages the MEMORY.md file by checking for duplicates, monitoring file size for optimal performance, and suggesting when a memory should be promoted to a high-priority project rule in CLAUDE.md.

Key Features

01Memory capacity monitoring with overflow warnings at 180 lines

02Manual memory anchoring for high-priority project discoveries

03Smart promotion logic to elevate memories to formal project rules

04Automated duplicate detection within existing memory logs

05Project-specific context isolation using local memory files

060 GitHub stars

Use Cases

01Storing team coding conventions and developer preferences that Claude should follow

02Documenting non-obvious debugging solutions like specific CLI flags for CI environments

03Recording project-specific architectural decisions such as library choices over alternatives

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add zhangzhang-111-i/claude-skills111 remember

For use in Claude.ai and ChatGPT

Key Features

01Memory capacity monitoring with overflow warnings at 180 lines

02Manual memory anchoring for high-priority project discoveries

03Smart promotion logic to elevate memories to formal project rules

04Automated duplicate detection within existing memory logs

05Project-specific context isolation using local memory files

060 GitHub stars

Use Cases

01Storing team coding conventions and developer preferences that Claude should follow

02Documenting non-obvious debugging solutions like specific CLI flags for CI environments

03Recording project-specific architectural decisions such as library choices over alternatives