Links <2026-04-16 Thu>
- robinovitch61/jeeves: AI agent conversation history browser
- calcom/cal.diy: Scheduling Software for Online Bookings
- JuliusBrussee/caveman: why use many token when few token do trick — Claude Code skill that cuts 65% of tokens by talking like caveman
- JuliusBrussee/cavekit: A Claude Code plugin that turns natural language into blueprints, blueprints into parallel build plans, and build plans into working software with automated iteration, validation, and cross-model peer review.
- affaan-m/everything-claude-code: The agent harness performance
optimization system. Skills, instincts, memory, security, and
research-first development for Claude Code, Codex, Opencode, Cursor
and beyond.
- Skills | ECC Selective Install Builder (SaaS around skills)
- forrestchang/andrej-karpathy-skills: A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.
- miolini/autoresearch-macos: AI agents running research on single-GPU nanochat training automatically adopted for MacOS
- karpathy/nanochat: The best ChatGPT that $100 can buy.
AI digest, June 2025
On <2025-06-11 Wed> I stumbled upon several articles on LLM I want to read later.
-
"How I program with Agents" by David Crawshaw (2025-06-08)
https://crawshaw.io/blog/programming-with-agents (local copy)
-
"How I program with LLMs" by David Crawshaw (2025-01-06)
-
"The Unreasonable Effectiveness of an LLM Agent Loop with Tool Use" by Philip Zeyliger (2025-05-15)
-
Sketch
- Sketch is a cloud-based and open source open source programming agent. It isolates work in containers to give agents more power and lets you work on many tasks in parallel.
- https://sketch.dev/
- Sketch is being retired, superceded by Shelley. Shelley is exe.dev's coding agent. boldsoftware/shelley: Shelley is a coding agent