# vCon Anthropic Chats

**Repo:** [vcon-dev/vcon-anthropic-chats](https://github.com/vcon-dev/vcon-anthropic-chats) · **First release:** May 2026

A standalone adapter that takes Claude AI conversation exports — the JSON files you get from the Claude web app's export feature, or from the API — and converts them into spec-compliant vCons.

## When to use it

* You want to put your team's Claude conversations into the same store as your call recordings, emails, and chats.
* You're building a corpus of LLM interactions for audit, training, or compliance.
* You need to apply [Lawful Basis](/extensions/lawful-basis.md) consent to AI conversations the same way you do to human ones.

## What the output looks like

The adapter uses the [Agent Session extension](/extensions/agent-session.md), so an exported Claude conversation produces:

* A `parties[]` array with the human user and one or more agent parties (each agent gets `role: "agent"` and a `meta.agent_session` block identifying the model and provider).
* A `dialog[]` array with each message as a text dialog entry.
* An `analysis[]` entry of type `agent_trace` containing the full session trace (tool calls, tool results, reasoning) as a JSON-encoded VAC document.
* Optional `attachments[]` for files generated or modified during the session (purpose: `agent_file_change`, `agent_artifact`, etc.).
* An `extensions: ["agent_session"]` declaration.

## Install and usage

See the repo README for the current CLI. The typical invocation:

```bash
vcon-anthropic-chats < claude-export.json > conversation.vcon.json
```

For batch processing, the same module is usable as a Python import.

## See also

* [Agent Session extension](/extensions/agent-session.md) — the spec the adapter produces against
* [vCon Adapter Development Guide](/vcon-adapters/vcon-adapter-development-guide.md) — patterns for building your own adapters


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://www.conserver.io/tools/vcon-anthropic-chats.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.