HostyAI

Auto-refresh (30s) —

Real-Time Trends

📊 Token Usage Over Time Hourly

⏱ Latency Over Time Avg ms

Activity Breakdown

📈 Requests & Errors Hourly

🏠 Top Properties by requests

Cost Analysis

💰 Cumulative Cost USD

🔬 Prompt vs Completion Tokens Per request

Breakdown by Model

🤖 Model Breakdown

📦 Compression Savings

Property Analytics

🏠 Top Property — Token Usage & Rush Hours

💰 Cost Per Property

System Health

🧠 Memory Retrieval Stats

🚨 Escalation Breakdown

⚠️ Error Types

📋 Status Distribution

System Observability

⚙ Circuit Breakers 0

Resilience state per external dependency. Open = short-circuiting calls; Half-open = probing.

Service	State	Failures	Opened at
Loading…

🎯 Agent Modes last 7d

Distribution of routing modes: refinery = new path, multi_agent = legacy, conversational = fast-path.

💵 Cost per Property last 7d

USD spent per property per day. Compare against the COST_PER_PROPERTY_DAY_USD_LIMIT cap.

Day	Property	Requests	Cost (USD)
Loading…

⚡ Refinery Stage Timings 0

Per-tool latency from recent data-refinery runs (parallel fan-out).

Loading…

🔧 Tool Usage by call count

🔍 Span Trace Explorer 0

Recent execution spans across all requests.

Loading spans…

🔍 Request Search 0

Time	Request ID	Conversation ID	Property Name	Property ID	Reservation ID	User ID	Endpoint	Latency	Prompt	Compl.	Total	Cost	Model	Status	Esc.

📄 Property Knowledge & Markdown Explorer —

Live v3 Hermes knowledge files, scoped reservation Markdown, and snapshot fields

Properties (0)

Loading properties from disk…

📄

Select a property or reservation on the left

Explore all verified Markdown files, guest details, and structured fields Hermes has access to.

property_data.md (Scoped to guest reservation if selected)

description.md (Listing description & neighborhood overview)

retrieval_snapshot.json (Complete raw API response with all fields)

💬 Conversation Browser —

Leave property empty to search globally by reservation ID, conversation ID, or guest query text.

Enter a Property ID and click Load Reservations

Select a conversation to load messages

Testing Lab — Test, Edit & Self-Learn

⚙ Test Configuration

User

Property

Reservation

Conversation

Or Custom Conv ID

Signature

💬 Conversations 0

Click Load to see conversations

🧠 Self-Learning Insights —

Click Refresh to load learning data

🔬 Chat & Messages 0

No conversation selected

Select a conversation or create a new one

          📈 Tokens
          │
          
            Last query
            —
          
          │
          
            Conversation
            —
          
          │
          
            Session
            ⇧0 ⇩0 Σ0

📷 Images

📝 Edit History (Self-Learning Data) 0

Click Load Edits to see self-learning data

📜 Live Log

Waiting for logs…

Auto-polling every 2s Auto-scroll

📄 Main Prompt Editor config.yaml

Eval Lab — LLM-as-Judge Evaluation & Prompt Feedback

📄 Test Suites

No suites yet

📜 Application Logs 0

No logs yet

📊 Token Usage Over Time Hourly

⏱ Latency Over Time Avg ms

📈 Requests & Errors Hourly

🏠 Top Properties by requests

💰 Cumulative Cost USD

🔬 Prompt vs Completion Tokens Per request

🤖 Model Breakdown

📦 Compression Savings

🏠 Top Property — Token Usage & Rush Hours

💰 Cost Per Property

🧠 Memory Retrieval Stats

🚨 Escalation Breakdown

⚠️ Error Types

📋 Status Distribution

⚙ Circuit Breakers 0

🎯 Agent Modes last 7d

💵 Cost per Property last 7d

⚡ Refinery Stage Timings 0

🔧 Tool Usage by call count

🔍 Span Trace Explorer 0

🔍 Request Search 0

📄 Property Knowledge & Markdown Explorer —

💬 Conversation Browser —

⚙ Test Configuration

💬 Conversations 0

🧠 Self-Learning Insights —

🔬 Chat & Messages 0

📝 Edit History (Self-Learning Data) 0

📜 Live Log

📄 Main Prompt Editor config.yaml

📄 Test Suites

🕑 Run History

📊 Case Results

💡 Prompt Improvement Suggestions

📄 Create Evaluation Suite

📜 Application Logs 0

hostyAI

📊 Token Usage Over Time Hourly

⏱ Latency Over Time Avg ms

📈 Requests & Errors Hourly

🏠 Top Properties by requests

💰 Cumulative Cost USD

🔬 Prompt vs Completion Tokens Per request

🤖 Model Breakdown

📦 Compression Savings

🏠 Top Property — Token Usage & Rush Hours

💰 Cost Per Property

🧠 Memory Retrieval Stats

🚨 Escalation Breakdown

⚠️ Error Types

📋 Status Distribution

⚙ Circuit Breakers 0

🎯 Agent Modes last 7d

💵 Cost per Property last 7d

⚡ Refinery Stage Timings 0

🔧 Tool Usage by call count

🔍 Span Trace Explorer 0

🔍 Request Search 0

📄 Property Knowledge & Markdown Explorer —

💬 Conversation Browser —

⚙ Test Configuration ↻ Reload Data

💬 Conversations 0

🧠 Self-Learning Insights —

🔬 Chat & Messages 0 🗑 Delete Conversation

📝 Edit History (Self-Learning Data) 0

📜 Live Log ALL INFO WARN ERROR ▮▮ Pause 🗑 Clear ▲ Hide

📄 Main Prompt Editor config.yaml ↻ Load 📚 Versions 💾 Save & Apply

📚 Prompt Versions ×

✎ Edit Message & Train Learning ×

📄 Test Suites

🕑 Run History

📊 Case Results

💡 Prompt Improvement Suggestions

📄 Create Evaluation Suite

📜 Application Logs 0

📋 Request Details ×

⚙ Test Configuration

🔬 Chat & Messages 0

📜 Live Log

📄 Main Prompt Editor config.yaml

📚 Prompt Versions

✎ Edit Message & Train Learning

📋 Request Details