ArchestackBest of Open-Source AI: List on GitHub ↗ (opens in a new tab)

#4242nd of 53 in Gateways and observability

claude-tap

Local proxy and trace viewer for inspecting AI coding agent API traffic

Documentation ↗ (opens in a new tab)Website ↗ (opens in a new tab)Repository on GitHub ↗ (opens in a new tab)
Stars
3.3k
License
MIT
Last commit
Oct 2026
Last release
Oct 2026
Language
Python

Missed the latest GitHub check; these facts may be older.

Overview

claude-tap wraps a coding-agent CLI (Claude Code, Codex CLI, Gemini CLI, Kimi, OpenCode, Pi, Hermes and others) in a reverse or forward proxy and records the real API requests. A browser viewer shows system prompts, message history, tool schemas, tool calls, reconstructed streaming responses, token usage and diffs between adjacent requests. Traces stay local and can be exported to a self-contained HTML file.

Who it is for: Developers debugging prompts, tools and context of AI coding agents

Strengths

  • Covers about 15 clients, including Claude Code, Codex CLI and App, Gemini CLI, and OpenCode
  • Structured diff between adjacent requests shows which prompt, tool or parameter changed
  • Traces stay on the local machine; common auth headers are redacted before recording
  • Exports a session to a self-contained HTML viewer for sharing or archiving

Weaknesses

  • Codex App capture needs a local CA trusted in the macOS login keychain
  • Native AWS Bedrock and Vertex require forward proxy mode; reverse mode breaks SigV4
  • Requires Python 3.11+ plus the client being traced; no Docker image
  • Tracing only; no routing, caching or cost control as in a full gateway

What it needs

  • no GPU
  • Needs Python 3.11+, The agent CLI to trace (e.g. Claude Code, Codex CLI)
  • Models: Anthropic API, OpenAI API, AWS Bedrock, Google Vertex AI, DeepSeek

Also in Gateways and observability

See all 53
Also in Gateways and observability
RankProjectScore
1OmniRouteLLM gateway routing one OpenAI-compatible endpoint across 372 providers86.8k stars+11,398 (+15%) in 30 daysMIT96 out of 100
2headroomCompresses tool outputs, logs and history before they reach the LLM79.5k stars+4,550 (+6.1%) in 30 daysApache-2.092 out of 100
3new-apiGateway that unifies LLM providers behind OpenAI, Anthropic and Gemini APIs51.8k starsAGPL-3.0known advisory in latest version91 out of 100
#4MLflowTracing, evaluation and prompt registry for LLM apps, plus ML tracking Live demo ↗ (opens in a new tab)28.8k starsApache-2.091 out of 100
#5LiteLLMOpenAI-format gateway and Python SDK for calling 100+ LLM providers63.2k starscustom license88 out of 100