# ai.getvda/traced-llm-proxy

> ai.getvda/traced-llm-proxy — ai-getvda-traced-llm-proxy. Use this tool when you need to intercept and modify interactions between language models and their clients, enabling logging, analysis, and debugging of large language model (LLM) requests and responses. It solves problems related to understanding and optimizing LLM performance, security, and compliance. The tool accepts LLM requests as input and outputs modified or logged responses, making it ideal for use cases involving model auditing, testing, and improvement.

Canonical page: https://skillsregistry.net/skills/ai-getvda-traced-llm-proxy  
JSON: https://api.skillsregistry.net/v1/skills/ai-getvda-traced-llm-proxy

## Description

Traced LLM Proxy

## Trust

- **Trust score (0–1):** 0.70
- **Verification tier:** unverified

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** mcp-remote
- **Runtime environment:** api
- **Updated:** 2026-09-03

## Source

- **Source listing:** [MCP Registry](https://registry.modelcontextprotocol.io/v0/servers/ai.getvda%2Ftraced-llm-proxy)

## Use it

MCP endpoint published by the skill: `https://traced-llm-proxy.getvda.ai/mcp`

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "ai-getvda-traced-llm-proxy"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/ai-getvda-traced-llm-proxy` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/ai-getvda-traced-llm-proxy/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt