slvDev
Featured
esp32-ai
A 28.9 million parameter language model runs on an $8 ESP32-S3 microcontroller entirely on-device, generating simple stories at about 9.5 tokens per second.
AI repository intelligence
FR-AI / ANALYSISWhy it is worth attention
It decouples domain knowledge from models, provides cryptographic integrity and audit trails, and offers strict verification modes to combat hallucination, all in a single Rust binary.
Who it is for
- Enterprise teams building internal AI assistants on sensitive documents
- Legal, compliance, insurance, fintech operations teams
- Developers needing local, version-controlled, auditable AI knowledge packages
- Organizations requiring GDPR compliance in AI knowledge systems
Use cases
- Creating a portable legal expert assistant that cites exact source spans
- Building a verifiable compliance Q&A system for insurance or fintech
- Serving a multi-package knowledge hub with package-as-model dispatch
- Exporting a package into Claude Code or OpenClaw harness layout
Strengths
- Portable single-file package (4MB Rust binary, no vector DB)
- Sentence-level provenance and claim lifecycle (extracted→reviewed→canonical→deprecated)
- Supports multiple LLM backends (Ollama, vLLM, OpenAI-compatible) without adapter
- Strict-render/verify modes and enforcement options (observe/warn/block) to reduce hallucination
Considerations
- No pre-built binaries yet (must build from source with Rust 1.75+)
- Enforce block mode is experimental, not a full hallucination elimination (only checks citation validity, not entailment)
- Security model is trust-on-first-use; no sandbox for MCP tools (treat untrusted .aipk like untrusted shell scripts)
README quick start
Quick Start
Description
AI assistant that cites reviewed source facts and flags what isn't covered — package your docs into one signed, portable .aipk file.
Related repositories
Similar projects matched by category, topics, and programming language.
jamesob
Featured
local-llm
A comprehensive guide for building and configuring a high-end local machine to run state-of-the-art LLMs, with detailed hardware choices, BIOS tuning, and Docker-based model serving.
makecindy
Featured
cindy
Cindy is an open-source AI agent that runs locally on your machine, integrates multiple AI harnesses and models, and provides memory, skills, and automation to perform real work in your projects and apps.