Agent OS · Component · ai-travel-concierge
Streaming travel concierge
What it does
Run a chat model with a hotel-search tool and return streaming recommendations and hotel cards.
Core pattern
Conversational intake + tool search and candidate cards
- Input
- Chat conversation
- Output
- Streaming answer and hotel candidates
Key parameters
| Name | Type | Default | Values |
|---|---|---|---|
| provider | string | — | — |
| messages | array | — | — |
| max_tool_steps | integer | 5 | 1–50 |
| allowed_tools | array | [] | — |
| message_history | — | — | — |
Maturity and limits
streamText route, Zod tools and chat components. Source paths were inspected statically; tests and live integrations were not executed. This rating describes implemented code, not production validation.
No root license found; verify reuse rights and media/font licenses. Provider credentials and customer/card information need protected handling.
Used in pipelines
Industry tags
- Travel & hotelsSource-supported industry
For work involving destinations, room rates and bookings, use conversational intake followed by tool search and candidate cards.
- Logistics & supply chainTransferable use
For delivery stops and shift scheduling, use conversational intake followed by tool search and candidate cards.
Skills pack
Contract scaffold and deterministic mock
Download · 12KBSHA-256 44cfb3985a7c8a4542e003969ab24ff0e9d17cae517b518b7ba092c9ec401eec
Related components
-
Versioned agent memory and knowledge lifecycle
Scoped memory retrieval + versioning and forgetting
Working 1 pipeline -
Contextual spiritual companion responses
Emotional cues and memory + supportive responses
Working 1 pipeline -
Contextual streaming assistant with history
Input and retrieval chaining + streamed replies and session history
Working 2 pipelines -
Trip intake and constraint extraction
Natural-language intake + hard constraints and missing fields
Working 1 pipeline