Dicklesworthstone

Dicklesworthstone / eidetic_engine_cli

Public

Durable, local-first, explainable memory for coding agents. A Rust CLI memory substrate built on the franken-stack (Asupersync, FrankenSQLite, Frankensearch, FrankenNetworkX).

10
0
100% credibility
Found May 04, 2026 at 10 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Rust
AI Summary

A single-binary command-line tool that creates durable, searchable, explainable memory for AI coding agents to recall project conventions, failures, rules, and evidence from past sessions.

How It Works

1
🔍 Discover the memory helper

You hear about a simple tool that helps AI coding buddies remember your project's rules and past mistakes so they don't repeat errors.

2
📥 Get it set up

With one easy download, you add the tool to your computer and point it to your project folder.

3
📚 Feed it your history

You let it scan old chat logs from your AI sessions to pull out useful lessons automatically.

4
💡 Ask for smart reminders

Before starting a task like fixing a release, you ask the tool what to watch out for, and it gives a neat summary of key rules and warnings.

5
📝 Save a new lesson

When you spot a helpful tip during work, you tell the tool to remember it for next time.

✅ Your AI never forgets

Now your coding assistant recalls project quirks, avoids old pitfalls, and works smarter every session.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 10 to 10 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 eidetic_engine_cli?

Eidetic Engine CLI (ee) is a Rust-built CLI that provides durable, local-first, explainable memory for coding agents. It solves agent amnesia by storing facts, rules, anti-patterns, and session evidence in a searchable substrate, then generating provenance-tagged context packs for prompts via commands like `ee context "fix release"`. Built on the franken-stack with Asupersync for runtime, Frankensearch for hybrid retrieval, and FrankenNetworkX for graph reasoning, it imports from tools like cass and runs offline.

Why is it gaining traction?

It stands out with explainable scores, procedural rules that decay on bad feedback, and graph boosts like PageRank—features absent in plain vector DBs or notes files. Developers hook it into Claude Code or Codex for task-specific context without cloud dependencies, getting deterministic JSON output and trauma-guard warnings first. The CLI-first design means no daemon needed for core workflows, unlike MCP servers.

Who should use this?

AI coding agent users building github durable functions, azure functions durable github workflows, or github durable task schedulers who need persistent recall across sessions. Ideal for teams maintaining github durable rules, streams, or task frameworks where agents repeat mistakes without memory. Suited for solo devs or small teams experimenting with local agents.

Verdict

Promising early tool for agent memory (10 stars, 1.0% credibility), with excellent docs, benchmarks, and e2e tests signaling maturity beyond its size. Try it if agent amnesia hits your workflow; skip for production without more adoption.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.