howardpen9

howardpen9 / hermes-gbrain-bridge

Public

Convert Hermes / OpenClaw agent memory (JSONL sessions, MEMORY.md) to markdown for gBrain ingest

agent-memory ai-agent bun claude-code codex-cli
12
0
100% credibility
Found Apr 13, 2026 at 12 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

A lightweight tool that gathers and converts conversation histories from various AI coding assistants into clean, searchable markdown notes for a personal knowledge base.

How It Works

1
🕵️ Realize your AI chat histories are scattered

You've used different AI helpers for coding, and their conversation notes are saved in messy spots on your computer, hard to search together.

2
📥 Pick up the handy bridge tool

You grab this simple tool that gathers and cleans up those notes into easy-to-use files.

3
🔍 Take a quick look at what you have

The tool scans your computer and shows you a list of recent chats from each AI helper, so you know what's ready to organize.

4
✨ Preview your cleaned-up memories

You see a sneak peek of your conversations turned into neat, private notes with sensitive info safely hidden.

5
💾 Save the notes to a folder

With one go, it creates simple text files of all your selected chats in a tidy folder.

6
📂 Feed them into your knowledge brain

You hand the folder to your personal knowledge tool, which absorbs everything for easy searching.

🎉 Search your entire AI history effortlessly

Now all your past coding ideas, decisions, and chats live in one searchable spot, ready whenever you need them.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 12 to 12 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is hermes-gbrain-bridge?

This TypeScript CLI tool, powered by Bun, converts agent memory from Hermes JSONL sessions and MEMORY.md files, Claude Code, Codex, and OpenClaw workspaces into clean markdown for gBrain ingest. It bridges fragmented formats like JSONL event streams and delimited markdown into one staging directory, letting you unify years of AI conversations into a single searchable brain. Run discover to scout files, export with mtime windows and size filters, then feed straight to gbrain import.

Why is it gaining traction?

Zero runtime dependencies mean instant setup—no npm drama. Built-in secret redaction, noise filtering (drops 90% junk from event streams), and a canonical markdown output keep embedding costs low, as proven in its case study slashing 1.4GB Claude Code to 200MB useful content for under $2. Developers dig the safe-by-default design: it never touches your DB, just spits staging files.

Who should use this?

AI agent enthusiasts chaining Hermes, Claude Code, Codex, or OpenClaw who run gBrain on Railway Postgres and hate siloed memory. Ideal for solo devs consolidating 30-day session histories or 1-year archives into semantic search, or teams wiring MCP tools like claude mcp add for brain access in every session.

Verdict

Grab it if your stack matches—it's a battle-tested MVP used daily, with stellar README docs and easy extension via adapters. 12 stars and 1.0% credibility score signal niche maturity, not broad polish, but the real-world metrics and principles make it reliable for agent-to-brain workflows.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.