llms.txt File

Machine-readable discovery file for AI agents. Helps Claude, LLMs, and AI applications understand Neatlogs capabilities, APIs, and integration options.

The llms.txt file is a machine-readable discovery document that helps AI agents, LLMs, and autonomous applications understand what Neatlogs can do and how to interact with it.

What is llms.txt?

Think of it as a manifest for AI agents. Just like robots.txt tells web crawlers what pages to index, llms.txt tells AI agents:

What APIs and tools are available
How to authenticate
What capabilities exist
Where to find documentation
Real-world use cases and workflows

When Claude or another AI agent encounters your product, it can read /llms.txt to immediately understand what it can do without having to read all your documentation manually.

Location

The file is served at the root of your domain:

https://neatlogs.com/llms.txt

Anyone can access it—no authentication required. It's public information about your platform.

What's Inside

The Neatlogs llms.txt contains:

Section	Content
Overview	What Neatlogs is and key capabilities
MCP Tools	All 15 available tools organized by category
Detections	What issues Neatlogs monitors automatically
Span Types	Execution types (LLM, TOOL, AGENT, etc.)
Authentication	How to get API keys and authenticate
API Info	Endpoints and base URLs
Use Cases	4 real-world workflow examples
Documentation Links	Pointers to guides and references
Rate Limits	Quotas and performance limits

Key Sections Explained

Overview

Brief summary of what Neatlogs does and who should use it:

Neatlogs is an observability platform for AI agents and LLMs. 
It captures execution traces from your agents, detects issues 
(hallucinations, regressions, anomalies), and provides AI-powered 
fixes with root cause analysis.

Available MCP Tools

All 15 tools listed with descriptions:

**Connection & Health:**
- ping - Health check
- whoami - Get authenticated project ID

**Query Production Data:**
- search_traces - Full-text search over traces
- get_trace_context - Fetch complete span tree
- list_detections - List active detection types
- get_detection_trend - Trend analysis for detections

[... and 8 more triage/fix tools]

Detections

What Neatlogs automatically monitors:

Quality Issues:
- hallucination - False or fabricated content
- prompt_regression - Quality degradation
- tool_contract_misuse - Tool called with wrong arguments

Performance/Cost Issues:
- cost_anomaly - Token usage/spend spike
- latency_anomaly - Response time degradation

[... more detection types]

Use Cases

Real workflows showing how AI agents use Neatlogs:

Debugging your own execution - Agent logs work, finds similar traces, compares
Self-healing via fixes - Agent reads suggested fixes, implements, updates status
Continuous monitoring - Agent tracks detection trends, escalates if worsening
IDE integration - IDE shows production traces and suggested fixes to developers

How AI Agents Use It

When Claude or another AI encounters your product:

1. Agent finds https://neatlogs.com/llms.txt
2. Reads the file to understand capabilities
3. Sees MCP endpoint and available tools
4. Learns authentication method (API keys)
5. Understands use cases and workflows
6. Decides it can help with observability tasks
7. Follows links to detailed documentation
8. Calls MCP tools via the API

Without llms.txt, the agent would need to:

Read all documentation manually
Guess at APIs and endpoints
Figure out authentication
Discover capabilities through trial and error

Format

The file is plain text with Markdown formatting:

# Neatlogs - AI Agent Observability Platform

## Overview
[Descriptive overview section]

## Primary Interface: MCP (Model Context Protocol)

### Endpoint
https://neatlogs.com/mcp

### Available MCP Tools (15 tools)

[Organized tool listing]

Why Markdown?

Human-readable by developers
Machine-parseable by LLMs
Standardized format
Easy to maintain

When It's Useful

llms.txt helps in these scenarios:

Scenario	How llms.txt Helps
AI agent exploring your platform	Immediate overview of capabilities
Claude using your API	Discovers MCP tools without manual lookup
IDE plugin integrating your service	Understands authentication and rate limits
Autonomous agent deciding if useful	Reads use cases to decide if relevant
Developer building on your platform	Quick reference of all tools and endpoints

Content Organization

The llms.txt follows a logical flow:

Overview - What is this?
Capabilities - What can I do?
Primary Interface - How do I interact?
Tools - What's available? (detailed list)
Detections - What gets monitored?
Documentation - Where do I learn more?
Use Cases - Show me examples
API Details - Technical specifics
Getting Started - How do I begin?

Updating llms.txt

When you add new features or tools:

New MCP tool added? → Update the tool list
New detection type? → Add to detections section
Documentation changed? → Update doc links
Rate limits adjusted? → Update quotas section
New use case discovered? → Add workflow example

Keep it current so AI agents get accurate information.

Best Practices

Do's:

Keep it updated with new tools and features
Link to accurate documentation
Include real use cases and workflows
Use clear, concise language
Explain authentication requirements
List all rate limits and quotas

Don'ts:

Include sensitive information (passwords, secrets)
Leave broken documentation links
Outdated tool or endpoint information
Unclear authentication instructions
Missing rate limit information

File Location

Development:

<project-root>/neatlogs-app/public/llms.txt

Production:

https://neatlogs.com/llms.txt

Raw access (no browser rendering):

curl https://neatlogs.com/llms.txt