INCIPIENT
Sandbox active v0.1.0 · TypeScript
Open-source coding agent · CLI + MCP Server

Every build
begins here.

INCIPIENT is a general-purpose coding agent that works through a transparent tool protocol: every action is visible, every change waits for your approval, and everything is recorded in an audit log. Try the agent loop yourself right on this page — no API key, and none of your real files are touched.

Built-in tools
7 tools
Approval mode
3 modes
Runtime deps
0 dependencies
Test suite
11/11 passing
INCIPIENT logo — a chrome hexagon with a lime glow on a dark indigo background
● SANDBOX| workspace: ~/my-project| endpoint: OpenAI-compatible
Live demo · In your browser

Interactive console

Give the agent a task, watch the tool blocks it sends, then decide for yourself: allow or reject. Switch the approval mode to feel how its safety rules change — including the sandbox that rejects paths outside the workspace and the automatic block on destructive commands.

Engine: faithful simulation Model: simulated Real files: untouched
Approval mode

~/my-project — incipient PLAN mode: ask
Example tasks — click to run
Free-form commands understood: create file X containing … · read X · search for … · list files · run <command> · remember: … · /mode ask|auto|readonly · /plan <task> · /memory
Capabilities

Designed like a professional tool

Not a chatbot bolted onto an editor — INCIPIENT is an agent with a protocol, a permission gate, and an audit trail that anyone can inspect.

01

Text tool protocol

The agent speaks through a single ```tool JSON block per turn — no proprietary function-calling. No tool block means the task is done. Runs on any OpenAI-compatible endpoint.

02

7 file & shell tools

read_file, list_dir, grep, write_file, edit_file, run_command, and remember — enough for a real read → change → verify cycle.

03

Approval-first

Reads are always free. Writes and commands ask for permission in ask mode, run immediately in auto — except destructive ones — and are always rejected in readonly.

04

Sandbox workspace

Every path is normalized and confined to the project folder. Escape attempts — even ../../etc/passwd — are rejected before the tool ever runs.

05

Project memory

Important facts live in .incipient/memory.md via the remember tool — read back at the next session, so the agent never starts from zero.

06

Full audit log

Every tool call — allowed, rejected, or blocked — is recorded with a timestamp in .incipient/audit.log. No silent actions.

How it works

One loop, three steps

The exact same agent loop you are running in the console above.

01

The model proposes a tool

The model takes your task and replies with plain text — or a single ```tool block holding the tool name and its JSON arguments. One turn, one action.

02

The safety gate decides

Before anything runs, the gate checks the approval mode, the sandbox path, and the destructive-command list. This is where you press Allow or Reject.

03

The result returns to the model

Tool output — file contents, grep results, test results — goes back as context for the next turn and lands in the audit log. The loop stops when the model replies without a tool block.

Safety model

The same rules, in every mode

This matrix is not theory — switch modes in the console and prove it on the same task.

Action typeaskautoreadonly
Read file · list · grep Free Free Free
Write / edit file Asks first Runs immediately Rejected
Ordinary shell command Asks first Runs immediately Rejected
Destructive command (rm -rf, mkfs, curl | sh…) Asked + warning Blocked outright Rejected
Path outside the workspace Sandbox-rejected Sandbox-rejected Sandbox-rejected
Quickstart

From the browser to your terminal

This page simulates the model. The real INCIPIENT is a TypeScript CLI + MCP server that connects the same loop to a real model through any OpenAI-compatible endpoint.

  • Zero runtime dependencies — TypeScript and Node.js only.
  • Two surfaces — an interactive CLI, or an MCP server (stdio) for other clients.
  • Identical safety — the same ask / auto / readonly modes, sandbox, and audit log you just tried.
terminal — incipientbash
# check that your environment is ready
$ incipient doctor

# give it a task — ask mode is the default
$ incipient run "add a /health endpoint and its test"

# or run it as an MCP server for other clients
$ incipient mcp