kolo - greppable debugger for Python agents

Python debugger built for AI agents: captures every executed function call, return value, local variable, and HTTP request into greppable trace files.

Saved Apr 28, 2026View source ↗

#observability #python #developer-tools

This entry doesn't have a long-form writeup yet. Follow the source link above for the full context.

Related entries

GitHub Tool

tokentap - intercept and visualize LLM token traffic

Python proxy that intercepts LLM API traffic and shows real-time token usage, costs, and context window in a terminal dashboard. Useful for debugging prompts and tracking spend.

#observability #cli #python #cost-tracking #developer-tools

GitHub ToolFeatured

Rapid-MLX - 2-4x faster local LLM inference on Apple Silicon

MLX-native inference engine with OpenAI-compatible API. The novel piece: DeltaNet state snapshots bring prompt caching to non-trimmable architectures (Qwen3.5 hybrids), restoring RNN state in ~0.1ms. 2-5x faster TTFT, native Metal kernels, continuous batching.

Why I saved this - DeltaNet state snapshots are described as the first prompt-cache technique for non-trimmable MLX architectures - Ollama and llama.cpp can't match this on hybrid RNN-attention models today.

#developer-tools #local-llm #apple-silicon #python #self-hosted

GitHub Tool

unclog - find and fix Claude Code context bloat

CLI that scans ~/.claude for MCPs, skills, hooks, and CLAUDE.md files, shows how many tokens each consumes before you type, and lets you remove items with reversible snapshots.

#claude-code #observability #context-engineering #cli #developer-tools

GitHub Tool

mcp-shark - Wireshark for Model Context Protocol

Electron capture and inspection tool for Model Context Protocol traffic. Records every HTTP request/response between an IDE and its MCP servers for forensic analysis.

#mcp #agent-security #observability #developer-tools #javascript