sim-moss

Moss semantic search integration for sim.ai workflows.

Provides MossSimSearch — a lightweight adapter that queries a preloaded Moss index and returns documents in sim.ai's expected {"content", "score", "source"} shape, ready to be served from a webhook that sim.ai calls as an external tool.

Installation

pip install sim-moss

Prerequisites

• Moss project ID and project key (get them from Moss Portal)
• Python 3.10+
• A sim.ai workspace with a deployed workflow

Quick Start

from sim_moss import MossSimSearch

search = MossSimSearch(
    project_id="your-id",
    project_key="your-key",
    index_name="my-docs",
)
await search.load_index()

result = await search.search("How do I reset my password?")
print(result.results)      # [{"content": "...", "score": 0.94, "source": "faq.md"}, ...]
print(result.time_taken_ms)  # 4

Wiring into a sim.ai Workflow

sim.ai workflows can call external HTTP tools. Point an HTTP tool node at a webhook server that wraps MossSimSearch:

POST /search
Body: {"query": "{{user_message}}"}

The response shape — {"results": [...], "time_taken_ms": 4} — maps directly to sim.ai's tool output. See examples/cookbook/sim for a complete FastAPI server and setup guide.

Configuration

MossSimSearch

Methods

SimSearchResult

License

This integration is provided under the BSD 2-Clause License.

Support

• Moss Docs
• Moss Discord
• sim.ai Docs