# AI and AI-assisted development

> Where the LLM enters the development loop: code review, static analysis mixed with models, deterministic filters instead of prompts.

- Record count: 5
- Source: https://muhammetsafak.com/labs/focus/ai-development/
- Language: en-US
- Author: Muhammet Şafak

---
In these experiments the language model gets one narrow job: reading a diff, drafting a fix from an error report, passing a RAG request through without holding a provider key. The experiments do not leave the model to the prompt alone; they run it alongside static analysis and deterministic filters. If you are looking for examples of AI-assisted development that stays checkable, the records here are for you.

[My AI-Assisted Engineering Workflow](https://muhammetsafak.com/engineering/essays/ai-assisted-engineering-workflow/) · [AI Didn't Remove the Bottleneck — It Moved It](https://muhammetsafak.com/engineering/essays/ai-didnt-remove-the-bottleneck/)

## All records

### TokenMetrics: making AI agent token spend visible per project

- A desktop app that reads Claude Code, Codex CLI and Gemini CLI logs locally and shows token usage by project, model and session.
- What it does today: Collects AI agent token usage from local logs and shows it by project, model, session and time, with estimated cost and subscription limit tracking. A developer who uses Claude Code, Codex CLI or Gemini CLI heavily and wants to see the spend can install it.
- State: In development
- Maturity: Beta
- Form: Open-Source, Desktop
- Started: 2026-11-08
- https://muhammetsafak.com/labs/tokenmetrics-lab/
- Markdown: https://muhammetsafak.com/labs/tokenmetrics-lab.md

### Serving the documentation of dozens of projects from one server

- Build notes on serving the documentation of many projects from one multi-tenant server, each with its own MCP endpoint and documents kept on your machine.
- What it does today: Makes a project's folder, git repository, Obsidian vault or Notion workspace searchable behind a single MCP endpoint, so an agent reaches the documentation without the data going to a third party. Anyone who can run Docker can host it and manage it from the dashboard.
- State: Became a product
- Form: Open-Source, Web
- Started: 2026-08-19
- https://muhammetsafak.com/labs/contextator-lab/
- Markdown: https://muhammetsafak.com/labs/contextator-lab.md

### Letting an application send RAG requests without ever holding a provider key

- A self-hosted Go gateway holds the provider keys and injects RAG context, so the application talks to one OpenAI-compatible endpoint.
- What it does today: Lets an application talk to one OpenAI-compatible endpoint while a project's configuration decides which model the request reaches and, when a document store is linked, which documents it draws context from. Anyone who does not want provider keys spread across applications, and wants spend capped per project, can run it on their own server.
- State: Became a product
- Form: Open-Source, Web
- Started: 2026-08-19
- https://muhammetsafak.com/labs/ragmux-lab/
- Markdown: https://muhammetsafak.com/labs/ragmux-lab.md

### SentryFixer: reading a Sentry error against the code and drafting the fix

- How a Claude Code skill reads a Sentry issue against the working tree and drafts the fix as a diff and a regression test, touching no file before approval.
- What it does today: Reads a Sentry issue against the working tree, produces a root-cause analysis and a fix plan, and touches no file without explicit approval. Anyone using Claude Code who needs to triage a production error can install the skill.
- State: Halted
- Form: Open-Source
- Started: 2026-07-01
- https://muhammetsafak.com/labs/sentryfixer-lab/
- Markdown: https://muhammetsafak.com/labs/sentryfixer-lab.md

### Local code review without letting the diff leave the machine

- How a provider-agnostic Go CLI reviews a diff without it leaving the machine, and how the experiment reached signed binaries as CommitBrief.
- What it does today: Reviews code locally, with the diff never leaving the machine; the `Provider` interface lets Anthropic, OpenAI, Gemini and Ollama plug into one contract. Signed binaries are published — anyone who does not want to hand their diff to a cloud provider can install it.
- State: Became a product
- Form: Open-Source
- Started: 2026-01-01
- https://muhammetsafak.com/labs/commitbrief-lab/
- Markdown: https://muhammetsafak.com/labs/commitbrief-lab.md