Peripheral FAQs

Question 1

What is Peripheral and what does it do?

Accepted Answer

Peripheral is a developer tool that builds a comprehensive self-model from your past AI conversations (e.g., Claude, ChatGPT). It extracts behavioral signals, distills them into executive (work style) and emotional (personality) constitutions, and serves this model as custom tools for other AI agents.

Question 2

How does Peripheral create my AI self-model?

Accepted Answer

It parses your conversation history to identify structured behavioral signals, distills these into executive and emotional constitutions, and then conducts AI-driven interviews across 10 psychological dimensions to fill any model gaps, ensuring an accurate and complete self-representation.

Question 3

Which AI platforms and models does Peripheral support?

Accepted Answer

Peripheral can parse conversation history from Claude Code sessions, Claude web chats, and ChatGPT exports. For distillation, interviews, and serving MCP tools, it supports both Anthropic (Claude) and OpenAI (ChatGPT) models, configurable via `peripheral.toml`.

Question 4

What are the 'MCP tools' and how do they use my self-model?

Accepted Answer

MCP tools ('My Current Persona') are functions that expose your self-model to other AI agents. These include `get_self_model` (full model), `get_style_rules` (writing style), `write_as` (ghostwriting in your voice), and `would_approve` (evaluating work against your preferences and values).

Question 5

Is my conversation data and self-model kept private?

Accepted Answer

Yes, your self-model data, including constitutions and generated files, stays entirely local to your machine within the `data/` directory, which is gitignored by default. Peripheral only makes API calls to your chosen model provider for processing; no data is sent or stored externally by Peripheral itself.

Peripheral

Peripheral

Key Features

Use Cases

Key Features

Use Cases