xepak
DSL based REST service for your Sqlite database (and not only) with LUA support, consider it as an alternative to PostgREST
Install / Use
claude mcp add rustrum -- npx -y github:rustrum/xepakIf the server publishes to npm under a different name, use that package instead — check the repo README.
MCP Server
Model Context Protocol server
Quality Score
Category
Data & AnalyticsSupported Platforms
Our assessment of xepak
xepak scores 81/100 on our quality scale, 521st of 609 Data & Analytics skills we index.
Its MCP Server is 6.9 KB long, well organised into 21 sections with 3 code examples: a thorough specification that gives an agent plenty to work with.
It has 10 GitHub stars, so there is little community track record yet; judge it on its content.
Maintenance, license and trust
- The repository was last updated today, so xepak is actively maintained.
- Our last check on 2026-11-13 found the source still online.
- It is released under the MIT license, a permissive license that allows use, modification and commercial use with attribution.
- Its trust signals score 97/100, with no cautions. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.
Safety scan
No issues foundOur scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful.
AI review by kimi-k2.7-code on 2026-11-13. Automated pattern scan on 2026-11-13. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.
xepak compared with similar skills
All 4 of these similar skills score higher than xepak; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| xepak (this skill)by rustrum | 81 | 10 | today | MCP Server |
| claude-memby thedotmack | 100 | 117.9k | today | CLAUDE.md |
| Agent-Reachby Panniantong | 100 | 112.7k | 10d ago | CLAUDE.md |
| rufloby ruvnet | 100 | 82.0k | today | CLAUDE.md |
| headroomby headroomlabs-ai | 100 | 79.9k | today | CLAUDE.md |
Frequently asked questions
- How do I install xepak?
- Run
claude mcp add rustrum -- npx -y github:rustrum/xepak. The install tabs above show the steps for each supported agent. - Which AI agents does xepak work with?
- It is written for Claude Code and Claude Desktop, as a MCP Server file. Other agents that read the same format can often use it too.
- Is xepak safe to use?
- Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful. It is MIT-licensed and scores 97/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
- Is xepak still maintained?
- The repository was last updated today, so xepak is actively maintained.
Skill content
View source on GitHubTL;DR
Imagine PostgREST but instead of Haskell with PL/SQL it is based on Rust with Lua and focused on Sqlite (and other DBs).
Summary
Xepak is a DSL configurable REST,JSON-RPC,MCP service that runs against SQL DB. Support SQLite for now with other DBs planned in the future.
Why focus on SQLite? Because it is amazing and extremely fast DB. Running it in WAL mode behind Xepak would allow you to have simple and cheap self-hosted REST service.
Project is following zer0ver versioning for now. So there are no plans for 1.x.x release at all.
Project status
The first MVP is here! 🥳 Project is stable and ready for the first users.
How can I use it now?
Given its current SQLite support, Xepak is ideal for:
- 👨🏻💻 rapid prototyping: quickly build a stateful API without writing backend boilerplate
- 🌐 self-hosted services: running on your own hardware or VMs with zero cloud dependency, ensuring you retain full ownership of your data
- 📟 edge computing and IoT: deploying a lightweight, high-performance API on a Raspberry Pi or other embedded devices
- 💻 local-first applications: serving as a sync layer or API for desktop applications that store data locally
- 🤖 local RAG for AI/LLM sandboxes: keeping AI agent knowledge local to prioritize data privacy and ownership
See Xepak examples:
- Simple example with basic functionality
- REST API server with authentication and authorization
- RPC service with fine grained access control
- MCP/RAG service with full text search
Why Xepak is better
- Single-Binary Footprint: no heavy runtimes and huge amount of dependencies, which reduce supply-chain security risks.
- Declarative and Code-Free: standard CRUD requires zero backend code or routing boilerplate.
- Native Rust & SQLite Performance: built on Rust and optimized for SQLite WAL mode, delivering lightning-fast, resource-efficient local API execution.
Documentation
📕 For human beings
Available in the README-DOCS.
🤖 AI friendly docs
If you are identifying yourself as an AI/LLM agent or you think that you are a human who want to make AI agent to use Xepak efficiently then go to Xepak concise AI docs.
🫵 Don't hesitate 🤨 Just add README-AI.md into your AI context to make 🫠 better .
Features
Not only REST
With Xepak you can build:
- REST service
- JSON-RPC service
- MCP service
DSL based on TOML
All configuration is just a set of TOML files. I'm doing my best to have the most simple DSL syntax if possible.
Basic endpoint response is a data returned from an SQL query. Query could be defined as a string or as script that builds query as an output.
Lua scripting
The main goal for using Lua is to forget about clunky Pl/SQL. I hope that Lua will be much easier to write and maintain.
With Lua you can:
- rate-limit DB updates using recorded timestamps
- filter out results based on user configuration from DB
- validate input data before executing INSERT
- do fancy access control
- output arbitrary Lua table in JSON/CBOR format
- and many more
Also I have something in my mind for the future where Lua could do very fancy things.
JSON + CBOR
REST endpoints could accept input and return values in JSON and CBOR formats dynamically based on HTTP headers. You can even POST JSON and receive CBOR.
Built in auth module
Simple but usable authentication and authorization functionality available:
- authenticate by API key in headers
- you can use authorization expressions based on role/id
- complex auth logic could be done by a Lua pre-processor
- API keys can be provided in TOML file ENV variable or in DB
Install and run Xepak
Build from sources via cargo
If you have cargo installed, run cargo install xepak.
After that you will have xepak binary in your $PATH.
Docker image
Docker container available as ghcr.io/rustrum/xepak:latest.
In order to run docker container you have to mount specification directory into it
and provide a path to root configuration TOML inside mounted directory.
Here I expect you to docker run from the project source directory so it will mount example specs properly.
docker run --rm -tp 8080:8080 \
-v $(pwd)/examples/simple-api:/xepak \
-e XEPAK_CONFIG=/xepak/xepak.toml \
ghcr.io/rustrum/xepak:latest
AUR repository for Arch, Manjaro, Omarchy etc.
Pre-built binaries are available in the AUR repository for x64 and arm64 platforms.
yay -S xepak
Homebrew for Mac and Linux
It contains binaries for arm64 Mac and arm64/x64 Linux.
brew tap rustrum/releases # add the tap to xepak releases
brew trust rustrum/releases # you can trust me for sure
brew install xepak # this will install the binary
Performance
I haven't run proper performance tests but you can look into simple API example which does provide simple tests scenarios for text queries and the same stuff wrapped in Lua scripts.
On a single core you can expect something like 500-900 req/sec. Two-four cores could boost performance to 2000-3500 req/sec.
Lua script makes things slower but not so much. The main bottleneck here is SQLite. Data roundtrip via Lua script without touching DB is much faster than running raw SQL queries.
License
This product distributed under MIT license BUT only under certain conditions that are listed in the LICENSE-TERMS file.
This project license will be always focused on personal private usage and small development teams. Maybe I will change the license in the future but the core principles would likely to stay.
Related Skills
claude-mem
117.9kPersistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More
Agent-Reach
112.7kGive your AI agent eyes to see the entire internet. Read & search Twitter, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu — one CLI, zero API fees.
ruflo
82.0k🌊 The original agent harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, federation, vector RAG integration, and native Claude Code / Codex / Hermes and many more Integrated
headroom
79.9kCompress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.
Languages
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.
