You're viewing a demo portfolio

Join the waitlist
PRSM

ai_filters_create

Active

Tool of io.github.saloprj/dialogbrain

declared in 1.17.0

Create a new AI filter for semantic intent-based message matching. AI filters use vector embeddings (via Voyage AI) to detect whether an incoming message matches a specific intent or topic. The filter's description is embedded as a reference vector at creation time. When a message arrives, its embedding is compared against this reference using cosine similarity. The description field is the most important part — it becomes the reference embedding that all incoming messages are compared against. Write it as a clear statement of what kind of messages should match: - 'Customer asking about pricing, subscription plans, or billing' - 'User reporting a bug, crash, or unexpected behavior in the product' - 'Inbound sales lead expressing interest in purchasing or trialing' The threshold controls sensitivity: 0.5 is a balanced default, lower values (0.3) cast a wider net, higher values (0.8) require closer matches. Note: This tool calls the Voyage AI embedding API to generate the reference vector.

Parameters schema

{
  "type": "object",
  "required": [
    "name",
    "description"
  ],
  "properties": {
    "name": {
      "type": "string",
      "description": "Filter name — a short, human-readable label (max 100 chars)"
    },
    "threshold": {
      "type": "number",
      "default": 0.5,
      "maximum": 1,
      "minimum": 0.1,
      "description": "Cosine similarity threshold for a message to be considered a match. Range 0.1–1.0. Default 0.50. Lower values (e.g. 0.3) are more permissive and catch more messages. Higher values (e.g. 0.8) require closer semantic similarity."
    },
    "description": {
      "type": "string",
      "description": "Reference text that defines what messages should match this filter. This text is embedded as a vector and used for cosine similarity comparison against all incoming messages. Be specific and descriptive — the quality of this text directly determines filter accuracy. E.g. 'Customer asking about pricing, subscription costs, or billing issues'. Max 500 chars."
    }
  }
}

What this tool wraps· 2 endpoints

min confidence0.700.50

Parent server

io.github.saloprj/dialogbrain

https://github.com/saloprj/dialogbrain-mcp

1/7 registries
View full server →
ai_filters_create — io.github.saloprj/dialogbrain — PRSM MCP