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.
AI repository intelligence
FR-AI / ANALYSISWhy it is worth attention
It offers a unique pipeline-driven reasoning flow (INITIAL → THINK → ANALYSE → OUTPUT), integrated tool calling (weather, shell, HTTP, etc.), agent handoff, and an interceptor system for logging/transforming messages, all within a Bun runtime.
Who it is for
- Developers building AI agents or chatbots with TypeScript
- Bun enthusiasts looking for a lightweight agent framework
- Engineers needing multi-agent orchestration with handoff support
- CLI power users who want interactive or one-shot agent commands
Use cases
- Answering questions that require live data (weather, exchange rates, Wikipedia)
- Executing shell commands programmatically with safety guardrails
- Delegating specialized tasks among multiple agents via handoff
- Building conversational agents with persistent memory across sessions
Strengths
- Pipeline-driven reasoning ensures step-by-step transparency
- Built-in tools cover weather, HTTP, finance, geo, shell, and more — no extra API keys for weather
- Safety guardrails on CLI commands (timeout, blocked patterns, output truncation)
- Extensible via interceptors and custom tools, plus optional long-term memory
Considerations
- Requires Bun runtime, not compatible with Node.js directly
- Only supports OpenAI (no other LLM providers)
- Shell command tool poses security risks; caution advised in production
README quick start
Installation
Related repositories
Similar projects matched by category, topics, and programming language.
yetone
Featured
kill-ai-slop
Kill AI Slop is a multilingual field guide and agent skill that catalogues 33 common AI-generated design tells with interactive before-after demos, and provides a tool to scan and fix web projects.
hahhforest
Featured
pi-textbook
A hands-on course that walks through building a Pi-style coding agent from scratch across 15 checkpoints, starting from an offline agent trajectory.