DalexorMI

Give Claude & Cursor aDeep Codebase Memory.

Every change you make is saved with a link to where it came from. You'll always know what changed, when, and why.

Dalexor runs quietly in the background without slowing your machine down or interrupting your flow.

Your full project history stays searchable forever. Nothing ages out, nothing gets deleted automatically.

Our servers run on redundant global infrastructure. Your history is available whenever you need it.

Three steps, and your AI assistant has a full memory of your project. No configuration file. No manual syncing. Just install, run, and ask.

Install the CLI tool with one pip command and rundalexor init. That's it — your machine is linked to your private account and ready to start watching.

Dalexor watches your files for meaningful changes. It ignores trivial edits (formatting, whitespace) and only saves what actually matters — with an AI-written plain-English summary of what changed and why.

Ask questions in plain English — right inside Claude or Cursor. Dalexor searches your entire saved history and returns the answer, usually in under a second.

"Migrated to JWT for stateless sessions..."

Built for developers using AI assistants.

Ask "why did we change the auth flow?" and get a real answer — not a list of files. It understands what you mean, not just what you typed.

Runs quietly in the background, watching your files as you code. You'll never notice it's there — but it's capturing everything that matters.

Every time something meaningful changes, AI writes a short, plain-English note explaining what happened. Like having a dev automatically documenting your own work.

Your code is encrypted before it ever leaves your machine. We can't read it. It's yours — fully isolated from other organisations.

See how your project has grown over time with a visual, scrollable timeline of every significant change — who made it, and what shifted.

Run as many projects as you want under one account. Search across all of them at once — useful when answers to today's bug live in a different repo.

Here's what actually happens every time you save a file — from your machine to your searchable project memory.

The change in your file is sent securely to our servers and queued for processing — takes milliseconds.

AI reads the change and writes a short explanation of what it does — in plain English, not just raw code.

The AI summary is converted into a format that can be searched instantly — even across thousands of past changes.

Your snapshot is saved permanently to your private account. It's available to search from any IDE, any time.

New devs can ask the project memory why things were built the way they were. Instead of bugging senior teammates with "why is this here?" — they just ask Dalexor.

>> "Why did we choose PostgreSQL?"
RECALL: "Switched from MongoDB for ACID compliance..."

When something breaks in production, you need to know when and why the relevant code changed. Dalexor finds that answer instantly — instead of combing through a year of git history.

>> "When did we add rate limiting?"
RECALL: "Added 2024-12-15 after DDoS incident..."

Dalexor writes documentation as you code. Every significant change gets an AI summary — so you always have an accurate, up-to-date record without writing a single doc manually.

>> "Show me all API changes"
RESULT: 47 Snapshots found...

If you ever need to prove what changed, when, and by whom — Dalexor has a permanent, searchable record of everything. Useful for audits, security reviews, or regulated industries.

Dalexor fits into how you already work. Install it, forget it's running, and let your AI assistant get smarter on its own.

Everything runs from the terminal. Check for conflicts, inspect your history, or start watching a new project — all without leaving your editor.

🔍 [DX VALIDATE] Checking conflicts...
[CRITICAL] High risk of merge conflict.

Prediction: "Recent activity in session_manager.py overlaps with your auth logic."

Log into your dashboard to browse your project's full timeline visually. See what changed, when, and how different pieces connect.

Dalexor works as a nativeMCP Serverinside Claude and Cursor. Your project memory is right there — no copy-pasting, no switching tabs.

- Use directly in Claude Desktop & Cursor
- "Find definition of UserSession?"
- "Predict conflicts for auth.py"
- Inline context without switching windows

Hook Dalexor into your CI/CD pipeline, Slack workflows, or any internal tooling via the REST API. Full programmatic access to your project's history.

Your code is private by design. We built every layer of Dalexor so that your data cannot be read, leaked, or shared — by us or anyone else.

All saved snapshots and history are encrypted using industry-standard ciphers. Your encryption keys are yours — we never hold them.

Your organisation's data is completely isolated at the database level. No one — including other Dalexor users — can ever see your project history.

The most sensitive parts run on your machine, not ours. Raw code is filtered and encrypted locally before anything leaves your device.

Dalexor doesn't just remember — it also warns you before things go wrong. It checks your current work against your team's recent activity and flags anything that looks like trouble.

Before you commit, Dalexor checks if anyone else on your team has recently touched the same files or functions. If there's a risk of collision, it tells you now — not after a messy merge.

Cross-references what you're changing now against what's already in your project history — so you don't accidentally undo a deliberate decision made six months ago.

Before you change a function, see exactly what else in the codebase depends on it. Know the full impact of your edit before you make it.

Runs quietly on your machine. Monitors your files, encrypts changes locally, and feeds context directly into your IDE — all without you doing anything.

Handles AI summarisation, history storage, and fast search across all your snapshots. Always available, globally distributed, and completely separate from your raw code.

Governed memory for coding agents with trust lifecycle, conflict detection, staleness tracking, and health scoring. SQLite + FTS5, zero infrastructure. Works with Claude Code, Cursor, Codex, Windsurf.

Agents Remember is a Drift-aware repository memory for coding agents in complex codebases. Captures what code can't say on its own! Retrieves memory by path, semantic search, and relationship (code-graph).

Persistent project memory + architectural decisions + pre-execution safety hooks for Claude Code. Local-only storage, multi-repo workspace, automatic knowledge extraction via background auditor.

MCP server that ingests project docs once and lets Claude search by meaning instead of reading everything — saving tokens on large codebases

Add persistent project memory and smart task routing to Claude Code. Stop repeating context every session - Claude finally remembers your architecture, decisions, and patterns forever. 100% local, 17 MCP tools.

Zero-config session continuity for Claude Code. Auto-captures context via Claude Hooks, provides 24 tools for memory, tasks, solutions, and knowledge graph. Multilingual semantic search (94+ languages).

Easily provide codebase context to Large Language Models (LLMs).

An intelligent codebase search engine that transforms local codebases into a natural language queryable knowledge base.

A local MCP server for AI coding agents. AST-aware indexing, semantic search, and automatic compression. Your agent stops re-reading your entire codebase every session.

A knowledge management tool for code repositories using vector embeddings, powered by a local Ollama service.