Projects
Open-source developer tools built for the AI agent era.
🔍 Cora CLI
AI Code Review, Your Way — CLI-first code review with BYOK.
| Language | Rust |
| Install | curl -fsSL https://raw.githubusercontent.com/codecoradev/cora-cli/main/install.sh | sh |
| License | MIT |
| Stars | |
| Version |
Review code with any LLM — OpenAI, Anthropic, Groq, Ollama, or any OpenAI-compatible API. Works in your terminal, CI/CD, git hooks, or directly inside AI coding agents.
Key features:
- Multi-LLM — Any OpenAI-compatible API, no lock-in
- Deterministic scanners — 11 security + 12 secret detection patterns, no LLM needed
- Pre-commit hooks — Catch issues before they reach CI
- SARIF output — Upload to GitHub Code Scanning
- Quality gate — Configurable pass/fail thresholds for CI enforcement
- MCP server — Expose rules to AI agents (Claude Code, Cursor, Copilot)
- Diff-hash caching — Skip repeat reviews automatically
- Custom rule engine — Write your own regex rules in
.cora.yaml
GitHub Action: cora-review-action
AI-powered code review as a GitHub Action. Runs Cora CLI on every PR, posts review comments automatically. BYOK, supports any OpenAI-compatible LLM.
GitHub → · Install from Marketplace →
🧠 Uteke
The Brain for Your AI — Offline-first semantic memory engine for AI agents.
| Language | Rust |
| Install | curl -sSL https://raw.githubusercontent.com/codecoradev/uteke/main/install.sh | sh |
| License | Apache 2.0 |
| Stars | |
| Version |
Single binary, zero dependencies, zero API keys. Your AI agent's memories never leave your machine.
Key features:
- Hybrid search — Vector (ONNX 768d) + FTS5 full-text, merged by Reciprocal Rank Fusion
- Rooms — Group memories by context (meetings, projects) with author attribution
- Time-travel queries — Recall memories as they existed at any point in time
- Relationship graph — Typed edges between memories (supersedes, contradicts, references)
- Smart decay — Composite importance scoring, pin critical memories
- Pluggable embeddings — ONNX (default), future OpenAI/Ollama backends
- Multi-agent namespaces — Fully isolated memory per agent, zero overhead
- 30ms recall — Library mode, ~42ms server mode
🪤 TrapFall
Self-Hosted Error Capture — Sentry-compatible, lightweight, and fast.
| Language | Rust + SvelteKit 5 |
| Install | docker pull ghcr.io/codecoradev/trapfall:latest |
| License | Apache 2.0 |
| Image size | |
| Version | Latest |
Drop-in Sentry alternative. Swap your DSN and errors flow to your own server. Real-time dashboard, Blake3 fingerprinting, webhook alerts — all in a 6MB Docker image.
Key features:
- Sentry-compatible ingest — Drop-in DSN swap, any Sentry SDK
- Multi-project support — Web, mobile, API — isolated DSN per project
- Real-time dashboard — SvelteKit 5 + Tailwind v4 + shadcn-svelte
- Blake3 fingerprinting — Automatic error grouping and deduplication
- Alert rules — Per-project condition-based webhooks with cooldown
- Full-text search — SQLite FTS5 + trigram substring search
- OpenAPI docs — Swagger UI at
/api/docs - MCP server — 12 AI agent tools via stdio JSON-RPC
- Tiny Docker image — 5.75MB (scratch + MUSL static binary)
💬 Rungu
Self-Hosted Feedback Board — Collect feature requests, bug reports, and roadmap priorities. Simple, fast, no SaaS lock-in.
| Language | Rust |
| License | MIT |
| Stars |
Lightweight self-hosted feedback collection. Give your users a place to share ideas and report issues — without depending on external SaaS services.
Key features:
- Feature request collection — Let users submit and vote on ideas
- Bug report tracking — Structured bug reports with status tracking
- Self-hosted — Your data, your server, zero external dependencies
