Quickstart

Choose how to connect your traces:

• Existing traces: connect Braintrust or Langfuse in Settings → Connections. This can import selected trace history and continue syncing according to the connector configuration.
• New events: use the Conversations SDK or MCP wrapper below, or send requests to the ingestion API.

This guide verifies data delivery. One event confirms that ingestion works; it is not enough by itself to define or validate an evaluator.

Before you startCopy Anchor Link

1. Get an API key

   In the Tensile dashboard, open Settings → API Keys and create a key. It looks like sk_ followed by a long hex string. Keep it out of source control. See Authentication for the full model.

2. Install an SDK

   Copy Text

   pip install honeai

   Copy Text

   npm install @asymmetric-ai/hone

Path A — Conversations SDKCopy Anchor Link

Use this when your agent runs as regular application code and you want to record each turn (the user's input and your agent's output). Wrap the turn with begin() and end().

import honeai

honeai.init("sk_your_key_here")

interaction = honeai.begin(
    user_id="user_8f21",
    agent_name="support-bot",
    input="How do I reset my password?",
)

# ... run your agent, produce a reply ...
reply = "Head to Settings, then Security, and choose Reset password."

interaction.end(reply, success=True)

Path B — MCP serverCopy Anchor Link

Use this when your agent exposes tools through a supported MCP server. The wrapper captures registered tool invocations; it does not currently capture resource reads or prompt calls.

import honeai.mcp
from mcp.server.fastmcp import FastMCP

server = FastMCP("inventory-tools")

# ... register your tools on `server` ...

honeai.mcp.track(server, api_key="sk_your_key_here")

server.run()

VerifyCopy Anchor Link

1. Trigger one turn (Path A) or call one tool (Path B).
2. Open the Tensile dashboard and go to Raw Logs — your call appears as a row with its operation name, status, latency, and timestamp.
3. Expand the row to inspect the input and output payloads.
4. For Path A, the turn may also appear under User Stories, grouped by the user_id you passed, when that workspace view is enabled for your organization.

If you see the row, the direct ingestion path is working. Evaluator discovery and pack review are separate workspace steps and depend on sufficient trace history and organization access.

Where to go nextCopy Anchor Link