Who used Claude on the Q3 audit files this week?
- search_messages
- list_sessions
Two people, three sessions. Each one links to its source log and byte offset.
Use case · Compliance and risk teams
Which staff used AI on this work, and can we prove it?
Staff use AI across the business. When someone asks who did what, on which work, you need a record, not a recollection.
Step by step
How it works for compliance. Each step is a real tool your AI calls.
- 01list_sessions
Sessions are recorded on each machine
An agent on each computer records Claude Code prompts, tool calls and file reads.
- 02get_compliance_activities
Chat activity is synced
For organisations with Anthropic's Compliance API, Claude chat activity, uploads and logins are synced in.
- 03get_session_detail
Every event keeps its address
The file and byte offset each event was read from is stored with it, so any answer traces to the line.
- 04get_cost_breakdown
Ask who, what and what it cost
Usage by person and project, spend over time, and a searchable audit trail.
Example
What it looks like. Sample data; the tool names are real.
Limits
What it doesn't do yet. So you know before you try it.
Need one of these for your team? Building it is the work Attri does.
Talk to Attri- Claude only today. Other AI platforms on request.
- It records and answers. It does not block prompts.
- Recording Claude Code needs the agent installed on each machine.
- No login or permissions in the open-source version.
Start with one tool. Connect it to your AI today.
Clone it from GitHub and add it to your MCP client. When your team needs it hosted, with login, talk to Attri.