Production monitoring · Autessa Evaluations

Conversation logs

Agent History keeps a full log of every agent conversation: the transcript, every tool request and response, response check results, the version, and the evaluation. It is AI agent observability built around the conversation, so every score traces back to the exact step behind it.

A core AI agent evaluation capability, done properly

Start free

Agent History · Billing support agentSAMPLE

Nov 3 · v4 · 6 messages · 2 toolsSuccessful

Nov 3 · v4 · 4 messages · 3 toolsSuccessful*

Nov 3 · v4 · 9 messages · 1 toolsUnsuccessful

Nov 2 · v3 · 3 messages · 1 toolsNot evaluated

⚙ refund_charge (2 messages)

input  { "orderId": "1042", "amount": 48 }
result { "status": "refunded" }

The problem

Scores without logs are a dead end

When a score drops, the next question is always “what did it actually do?” If the answer needs a separate tracing tool, a log export and a spreadsheet, nobody looks. Here the scores sit next to the full conversation.

What you get

Pick an agent and a date range, and optionally include your own sandbox test conversations.
Each row shows its evaluation status at a glance, with the date, version, message and tool counts.
The conversation viewer has an overview card: conversation ID, close date, agent version, API key, message and tool counts, and language for multilingual agents.
The full transcript includes audio transcriptions and attached-image markers.
Tool call cards group each request and response, with input variables, the result (raw or pretty JSON) and debug logs.
Response checks cards show what each check found and what happened to the reply.
An Evaluation Details panel shows the automated scores and lets reviewers add a human evaluation.

How it works

How does conversation logs work in Autessa Evaluations?

Open Agent History
From Agentic Workflows, choose the agent and the date range.
Scan the rows
The icon on each row shows successful, successful with caveats, unsuccessful or not evaluated.
View details
Read the transcript and expand any tool card to see its exact inputs and outputs.
Open Show Evaluation
Read the scores, and record a human evaluation if needed.

Every option

Filters and fields

Option	What it does
Select Agent	Owned, shared and company agents.
Start Date / End Date	Defaults to the current month.
Include My Sandbox Conversations	Show your own test runs alongside real traffic.
Overview card	Conversation ID, close date, version, API key, message and tool counts, language.
Tool call cards	Input variables, result with raw / JSON toggle, debug logs.
Evaluation Details	Automated scores and human evaluation with notes.

From code

Do it from the CLI or API

Run `autessa <command> help` for the full list of arguments.

CLI documentation →

Read history from the CLI

autessa agents conversations history --agentId <agentId> --page 0

autessa agents conversations conversation \
  --agentId <agentId> \
  --conversationId <conversationId>

Permissions

Who can do what?

Who	Can
The person a conversation ran as	See conversations started through the API or an app as them.
Agent Admins and company Admins	See every conversation for the agent.
Write access	Record human evaluations from the viewer.

Connected

How it connects to the rest of Evaluations

EVALUATIONS FEATURE

Human evaluation

Label conversations right where you read them.

EVALUATIONS FEATURE

Export and sync

Take logs and evaluations into your own analysis.

AUTESSA PLATFORM

AI Agents

Every agent, in every channel, writes to the same history.

FAQ

Questions about conversation logs

Does Autessa log tool calls?

Yes. Every tool request and response is grouped into a card with the input variables, the result and debug logs.

Who can see a conversation?

Conversations started through the API or an app are visible to the person they ran as, to the agent’s Admins and to company Admins.

Are test conversations mixed in with real ones?

Only if you ask. Tick Include My Sandbox Conversations to see your own test runs alongside real traffic.

Also in Production monitoring:Human evaluation and labeling Version performance Quality dashboards Multi-agent evaluation Export and sync

Try conversation logs in Autessa Evaluations

Agent History logs every conversation in full, with each tool request and response, so any score can be traced to the exact step behind it.

Start free

Test sets · simulated users · four-dimension scoring · response checks · grading API

100% customizable, code and all

Make Autessa Evaluations fully your own.

Use it out of the box, then change anything. Ask Autobot, Autessa’s AI builder, for a new field, screen, report, workflow or agent and it builds it into your copy of the app. Or open the code and edit it yourself. Nothing is locked: the app, its data and its code are yours.

Start free and make it yours