Langfuse
Open-source tracing, evals, and prompt management for LLM apps
About Langfuse
Langfuse is the observability layer most LLM apps are missing. It traces every call your app makes — prompts, completions, tool calls, latency, token cost — and gives you evals, datasets, and prompt management on top, so you can answer "why did it do that?" without re-running the whole thing by hand.
What you get
- Tracing — nested spans across chains, agents, and tool calls, with cost and latency attached to each step.
- Evals — run model-based or custom scorers over datasets so prompt changes get measured instead of vibed.
- Prompt management — version prompts outside your codebase and roll them forward without a deploy.
- Integrations — OpenTelemetry, LangChain, LlamaIndex, LiteLLM, and the OpenAI SDK, so instrumenting is usually a few lines.
Good for
Anyone running an LLM feature in production, or anyone whose API bill is growing faster than their understanding of it. Self-host the whole thing or use the managed cloud; the core is MIT-adjacent open source and the project came out of YC W23.
Repo: github.com/langfuse/langfuse.
More Github
Sepia
Sepia is a de-AI writing skill for any Agent Skills-compatible agent, with native plugins for Claude Code, Codex, Grok Build and Antigravity. It applies rules matched to where the text is going for professional prose, and narrative repair for fiction. It is based on the StoryScope paper (arXiv:2604.03136). Useful for anything you publish under your own name: docs, launch posts, newsletters. Install in Claude Code: /plugin marketplace add Nanako0129/sepia MIT licensed. View on GitHub .
Tastemaker
Tastemaker is an agent skill for Claude Code that grounds AI-generated UI in real reference images and a persistent, per-developer taste profile, instead of the generic defaults that make every generated screen look the same. Install in Claude Code: /plugin marketplace add codeswithroh/tastemaker MIT licensed. View on GitHub .
Worktrunk
Worktrunk is a CLI for Git worktree management, designed for parallel AI agent workflows. Each agent works in its own worktree, with its own branch and its own folder, so two sessions never edit the same checkout. If you run Claude Code, Codex or another agent on more than one task at a time, this is the plumbing that keeps their changes apart until you review and merge them. Written in Rust. Docs at worktrunk.dev , source on GitHub .
Birdview
Birdview maps your codebase architecture before your AI coding agent makes a change, so edits happen with the whole system in view instead of only the file in front of the agent. It is aimed at the most common failure of agent-written code: a change that is locally correct and globally wrong, because the agent never saw how the pieces connect. Open source under the MIT license. View on GitHub .
An agent skill that stops your AI generating generic, obviously-AI interfaces
taste-skill
AI coding agents converge hard on the same visual defaults: the same gradients, the same card layouts, the same spacing. taste-skill is a skill you load into your agent to push it away from those defaults and toward design decisions that look deliberate. Worth trying if everything you ship looks like it came out of the same template, because to a large extent it did.