2025-02-25 Tue

#AI Coding
- Had a fairly substantial #Agentic Vibe Coding moment, due in part to Person/Harper Reed/Blog/25/02/My LLM codegen workflow atm and Person/Geoffrey Huntley/Blog/25/02/You are using Cursor AI incorrectly.... The new Claude 3.7 Sonnet models really shine in CursorAI when you combine CursorAI/Project Rules with a spec.md, prompt_plan.md and a todo.md.
  - "Tools will come / Tools will go / Only the vibe coder / remains" - #Quote from Person/Rick Rubin 2025-02-22 Sat
#Big News
- Anthropic/Blog/25/02/Claude 3.7 Sonnet and Claude Code
  - my #notes
    Anthropic is possibly differentiating themselves by adding something kind of like a Continuously Variable Transmission for model reasoning, rather than having to pick between a reasoning model and a traditional non-reasoning model.
    API users can set a budget, which will be essential for controlling reasoning
    control LLM Reasoning Model budget through Anthropic/API - #Cool
    through the API, users can also control the budget for thinking: you can tell Claude to think for no more than N tokens, for any value of N up to its output limit of 128K tokens. This allows you to trade off speed (and cost) for quality of answer
  - There's a reason Claude is the default model for all Cursor users—Anthropic's approach to building models delivers on real-world tasks. During our extensive testing of Claude 3.7 Sonnet, we've seen significant improvements in the model's ability to understand and handle complex codebases and multi-step tasks. Now with two ways to think, Claude cements its place as the industry leader for coding. -- #Quote from Person/Michael Truell CursorAI Cursor, CEO
  - Anthropic/YouTube/25/02/Introducing Claude Code #Filed
    - my #notes
      One of the first times I've seen a compelling Chat #UI in #CLI - is this Curses?
      will this remove the need for tools like Repomix ?
#Filed
- Least Recently Used Cache
#Ideas
- StreamIO/Chat/Idea/MCP Server for LangGraph Client #Filed
#News
- CursorAI/v/0.46 - Agent is ready and UI Refresh #Filed
  - CursorAI/Security
    Ignore files: CursorAI/.cursorignore now blocks files from being added in chat or sent up for tab completions, in addition to ignoring them from indexing.
    NEW: CursorAI/.cursorindexingignore for specifically controlling file indexing
    TODO #Q when would I want to put something in .cursorindexingignore instead of .cursorignore?
    ... if I wanted something to be available to cursor in file references but not indexed? What's the use of that?
    perhaps the use case is that you have confidential information committed to the repo that's not a secret
  - #MCP improvements
    Agent can now automatically run MCP tools with Yolo mode
    Configure Project Servers with <project-root>/.cursor/mcp.json #Awesome
    TODO what's the json schema for this #q
    Support for Agent to use MCP/Resource as context
    I didn't know that Agent didn't support these ... this is essential to CursorAI/Idea/Proj Rules GitHub CLI AI SLDC/MCP
People
- Person/Michael Truell Cursor CEO
Research
- GitHub/Project/Concept/Terminal Status
  - Best Practices
    Use status to track workflow state, not completion
    For completion tracking, use:
    The issue's closed/open state
    Milestones
    Labels
    Consider automating status updates based on issue events
#TIL
- ChatGPT/Projects can be color coded
#Tips
- CursorAI/Technique/Logseq/Summarize Meetings
Too Bad, So Sad
- hit my monthly limit for CursorAI/Fast Requests. Really noticed it in the afternoon. Turns out, the CursorAI/Status Page said they were encountering rate limiting issues with Claude 3.7 Sonnet.
  - Monitoring - With the release of Claude 3.7, demand to try the new model is currently causing some rate-limiting issues, where users may see connection errors when attempting to use Claude models. To help mitigate this high demand, we have temporarily removed Claude 3.7 from free trial and slow pool users, while we resolve the usage issues.
    Feb 25, 2025 - 16:47 UTC
  - See CursorAI/Requests for details about fast and slow request pools
  - See CursorAI/Usage-Based Pricing for pricing beyond limits
  - See CursorAI/Settings/Models for model-specific limits

1 Linked Reference

Unlinked References