md-sentry

Q: How to install md-sentry?

Create the skills directory: mkdir -p .claude/skills. Then run: mkdir -p .claude/skills && curl -o .claude/skills/md-sentry.md https://raw.githubusercontent.com/yeet-src/md-sentry/main/SKILL.md. Finally, /md-sentry in Claude Code.

Q: What is md-sentry best for?

md-sentry is categorized under Documentation. It covers: agent.

New

1GitHub TrendingDocumentationby yeet-src

eBPF integrity monitor for an LLM agent's markdown brain: CLAUDE.md, skills, memory. Tags AGENT vs EXTERNAL edits.

First seen 6/12/2026

Summary

md, skills, memory) and tags whether changes come from the LLM agent or external sources.

This helps developers maintain awareness of who modified their agent's configuration and knowledge base, preventing unauthorized or accidental changes.

Install & Usage

Create the skills directory

mkdir -p .claude/skills

Download the skill file

mkdir -p .claude/skills && curl -o .claude/skills/md-sentry.md https://raw.githubusercontent.com/yeet-src/md-sentry/main/SKILL.md

Invoke in Claude Code

/md-sentry

Use Cases

Monitor when CLAUDE.md is modified by the agent vs. manually by the developer to ensure alignment with intended behavior.

Detect external tampering with skill files that could alter the agent's capabilities without the developer's knowledge.

Audit changes to memory files to understand what information the agent has stored or forgotten over time.

Track edit provenance during collaborative development to see which team members or processes modified the agent's configuration.

Integrate with CI/CD pipelines to flag unexpected modifications to agent configuration files before deployment.

Debug unexpected agent behavior by reviewing the history of changes and their sources.

Usage Examples

/md-sentry status — Check current integrity status of all monitored markdown files.

/md-sentry log --since 1d — Show all edits from the last 24 hours with AGENT vs EXTERNAL tags.

Monitor CLAUDE.md for external changes and alert me if someone edits it outside of the agent.

View source on GitHub

agent

Security Audits

LicenseUnknownSourceWarnRepositoryPass

Frequently Asked Questions

What is md-sentry?

md-sentry is an eBPF integrity monitor that tracks edits to your markdown brain files (CLAUDE.md, skills, memory) and tags whether changes come from the LLM agent or external sources. This helps developers maintain awareness of who modified their agent's configuration and knowledge base, preventing unauthorized or accidental changes.

How to install md-sentry?

To install md-sentry: create the skills directory (mkdir -p .claude/skills), then run: mkdir -p .claude/skills && curl -o .claude/skills/md-sentry.md https://raw.githubusercontent.com/yeet-src/md-sentry/main/SKILL.md. Finally, /md-sentry in Claude Code.

What is md-sentry best for?

md-sentry is a skill categorized under Documentation. It is designed for: agent. Created by yeet-src.

What can I use md-sentry for?

md-sentry is useful for: Monitor when CLAUDE.md is modified by the agent vs. manually by the developer to ensure alignment with intended behavior.; Detect external tampering with skill files that could alter the agent's capabilities without the developer's knowledge.; Audit changes to memory files to understand what information the agent has stored or forgotten over time.; Track edit provenance during collaborative development to see which team members or processes modified the agent's configuration.; Integrate with CI/CD pipelines to flag unexpected modifications to agent configuration files before deployment.; Debug unexpected agent behavior by reviewing the history of changes and their sources..