HN Showcase
Back to browse
GitHub Repository

Enrich coding agent prompts with codebase context using a local LLM

33 starsTypeScript

Promptscout a local prompt enricher for Claude Code

by obsfx·Feb 13, 2026·2 points·0 comments
Visit ProjectView on HN

AI Analysis

●●SolidSolve My ProblemWizardry

Local LLM context injection for Claude Code beats manual file discovery, but niche audience.

Strengths
  • •Local-only design with quantized Qwen 3 4B avoids API costs and token overhead
  • •Claude Code plugin integration hooks prompt submission flow transparently
  • •Leverages ripgrep + git for fast, accurate codebase retrieval without expensive indexing
Weaknesses
  • •Narrow use case: only valuable if already using Claude Code daily on large codebases
  • •~3GB model download and C++ compilation barrier limits casual adoption
Target Audience

Claude Code users, developers using coding agents on private codebases

Similar To

Continue.dev · GitHub Copilot context window optimization · Cursor's codebase indexing

Post Description

I use Claude Code daily and got tired of repeatedly listing relevant files in prompts. So I built promptscout: a small local tool that takes your prompt + repo file tree and enriches the prompt with the most relevant files/paths, so Claude has a better starting point.

It uses a quantized Qwen 3 4B model and runs fully on your machine. I also added a Claude Code plugin via UserPromptSubmit that calls the CLI and swaps in the enriched prompt before the request goes to Claude.

Similar Projects

AI/ML●●Solid

Experience-engine – reflection-based memory layer for local LLMs

Turns chat history into structured 'belief' and 'cognitive pattern' blocks you can inject into prompts, with simple APIs like run_reflection and run_synthesis that read like a research prototype. It's smart about separating V1 (domain beliefs) from V2 (transferable cognitive patterns), but it's clearly early-stage — tiny repo, Ollama-only workflow, and few commits mean you should treat it as an experimental MVP rather than a drop-in production memory system.

Big BrainNiche GemShip It
ashishluthara
313mo ago