-
Queries
-
Tokens
-
Tok/Q
-
Sessions
-
Users
-
Avg Lat
-
Max Lat
-
Good
-
Bad
-
FT Data
-
Cost
| Mode | Queries | Tokens | Tok/Q | Sessions | Users | Avg Lat | Good | Bad | Cost |
|---|
Loading...
100%
Logs
[Ready]
Loading...
Add New User
User Tiers Management
Users in Tier
Loading...
🤖 Agent API Keys
API keys for LLM providers used by agents
Not configured
Not configured
Not configured
Not configured
🔍 Search & Grounding
API keys for web search and grounding features
Not configured
Not configured
💾 Storage & Logging
Configure where logs and data are stored
BigQuery Active
Connected
Not configured
🔗 Deployed Agent Endpoints
API endpoints for deployed flows - use these in your applications
re::dact
POST /api/v1/agents/redactre::check
POST /api/v1/agents/recheckre::search
POST /api/v1/agents/researchre::doc
POST /api/v1/agents/redoc🔑 API Access Keys
Generate API keys for external access to your agents
Production Key
rk_live_****************************Development Key
rk_test_****************************🔐 Custom Secrets / Environment Variables
Define custom secrets for use in flows and agents
-
Today Total
-
Accepted
-
Rejected
-
Errors
-
Avg Quality
-
Sources
-
Nano Cost
Languages
📁All Sources0
-
Total Archived
-
Today
-
Errors
-
Avg Quality
-
Running
Languages
📁All Sources0
Archive Source
Add New Source
📝 Phase 2 System Prompt
Ten prompt jest używany przez LLM w Phase 2 (Verification) do oceny relevance fragmentów artykułu.
Zmienne: {title}, {description}, {chunk}
🧠 V2 Phase 2 System Prompt
Ten prompt jest używany przez LLM (Mistral/Gemini) w Phase 2 do wyciągania cytatów z artykułu. Prompt musi zwracać JSON z matches.
Scraper Settings
Rate Limiting
LLM Configuration
Storage
Bot & Mode Configuration
Article Preview
-
Live
-
Archive
-
App Logs
-
Uploaded
-
Total
-
Tokens
BigQuery Datasets
Loading...
🧪 Dataset Recipe Builder
📰 Live:
30%
📦 Archive:
30%
💬 Logs:
20%
📤 Upload:
20%
📋 Recipes
No recipes yet
📤 Uploaded Files
No files uploaded
Upload Training Data
Expected: {"text": "..."} or {"input": "...", "output": "..."}
🧠 Training
OFF
0
MODELS
0
JOBS
$0
TODAY
Current Production Model
No production model set
Trained Models
No models trained yet
Training Jobs
No active jobs
✨ Distillation
OFF
0
TOPICS
0
GENERATED
0
APPROVED
Topics
No topics. Add one to start generating.
Generation Config
Select topics to see estimate
Recent Texts
No synthetic texts yet
⚡ v1: Embedding
Local e5 filter → Cloud embeddings → Mistral
🏠 LOCAL Pre-filter (FREE)
Xenova/multilingual-e5-base
BROWSER
Runs 100% locally in browser via transformers.js
Pre-filters irrelevant pages → saves cloud API costs
☁️ Phase 1: Cloud Retrieval
0.45
🎯 Phase 2: Verification
0.1
0.70
🧠 v2: Two-Phase LLM
Local Qwen3 filter → Mistral quote extraction
🏠 Phase 1: LOCAL Filter (Qwen3)
✅ Auto-downloads (transformers.js)
Model: Qwen3-0.6B-ONNX (thinking mode)
Task: YES/NO relevance with reasoning
💰 FREE - runs 100% in browser, cached per URL+project
☁️ Phase 2: Quote Extraction
Research Assistant prompt:
Szuka: cytatów ekspertów, unikalnych tez, kluczowych danych
Szuka: cytatów ekspertów, unikalnych tez, kluczowych danych
How v2 works:
1️⃣ Qwen3 (local, FREE) - myśli i decyduje: relevant?
2️⃣ If YES → Mistral (cloud) wyciąga dokładne cytaty
3️⃣ Cache URL+project = instant repeat visits
1️⃣ Qwen3 (local, FREE) - myśli i decyduje: relevant?
2️⃣ If YES → Mistral (cloud) wyciąga dokładne cytaty
3️⃣ Cache URL+project = instant repeat visits
Toggle ON to enable v2
📡 Page Monitor
📋 Active Config
v1
Loading...