Agen-IT
Open Portal →
AI · ASPICE V-MODEL · EMBEDDED

An AI co-pilot that
grows with you

Agen-IT remembers your code style, decisions, and project context. Persistent soul + per-project memory, auto-activating skills, and a disciplined V-Model pipeline you can trust.

Launch Portal →See what it does
92+
Slash commands
16
Built-in skills
SWE.1–6
V-Model stages
5
Squad agents

/ Features

Built for real engineering

Eight pillars that turn an LLM into a disciplined engineering teammate — not a chat window.

Memory

Profile-Driven Co-Pilot

Onboards once, remembers your stack, standards, and decisions. Soul + per-project memory keep context across every session — no more repeating yourself.

Process

ASPICE V-Model Pipeline

Disciplined SWE.1 → SWE.6 lifecycle: requirements, architecture, implementation, V&V, and traceability audit — with explicit human approval gates.

Skills

Auto-Activating Skills

16 built-in Gemini CLI skills plus Pocock-adapted engineering skills (/diagnose, /tdd, /triage) activate by intent — no manual wiring.

Agents

Mission Squad

A parallel agent fleet — Archie, Developer, Tester, Analyzer, Debugger — coordinated through a steering queue for multi-turn tasks.

Hardware

Hardware-in-the-Loop

First-class JLink, CAN/LIN, and Saleae Logic 2 integration. Arduino-as-HIL turns dev boards into test nodes for real embedded validation.

Extensibility

Marketplace & MCP

Unified marketplace for MCP servers, Claude skills, and Gemini extensions. Compose your toolchain without forking the orchestrator.

Spec

Spec-Driven Development

Nine-phase SpecKit pipeline takes you from constitution to issues, with PMO standards enforcement and budget tracking built in.

Autonomous

Overnight Autonomy

/overnight runs a git-backed Plan → Code → Test → Evaluate loop with worktree parallelism and per-iteration rollback safety.

/ Architecture

Three layers. Clear contracts.

A stateful TypeScript orchestrator drives Gemini CLI for LLM work, which delegates to a Python bridge for hardware and heavy parsers.

1
Layer 1

TypeScript Orchestrator

Node ≥ 20 · React Ink
  • ▸REPL, slash commands, autocomplete, history
  • ▸V-Model state machine + approval gates
  • ▸Soul / project memory (markdown, file-locked)
  • ▸Mission Squad fleet · /goal autonomous loop
  • ▸Codedigest BM25 indexer
2
Layer 2

Gemini CLI

OAuth · No API key
  • ▸Auto-activates SKILL.md files by intent
  • ▸Built-in tools: read_file, write_file, run_shell_command
  • ▸Hooks: BeforeTool / AfterTool / Model
  • ▸MCP server orchestration
3
Layer 3

Python Tool Bridge

Python 3.11+
  • ▸pylink-square (JLink) hardware access
  • ▸cantools + python-can (CAN/LIN buses)
  • ▸Saleae Logic 2 API
  • ▸pdfplumber · openpyxl · python-docx

/ V-Model Pipeline

From requirement to audit

Each stage activates the matching skill, writes artifacts to project memory, and stops at an approval gate you control.

01
/req
SWE.1 Requirements
Parse customer specs into EARS / user stories
02
/arch
SWE.2/3 Architecture
HLD, components, interfaces, diagrams
03
/code
SWE.4 Implementation
Profile-aware code with traceability tags
04
/test
SWE.5 V&V Testing
Unity / CppUTest / Jest / Playwright per profile
05
/audit
SWE.6 Audit
Traceability + compliance audit (ASPICE / OWASP)

/ Live Demo

Try the REPL

A mock of the Agen-IT REPL so you can feel how it works before installing. Type a slash command or anything in free form — outputs are simulated, but the flow is the real one.

  • ▸ 92+ slash commands · this demo wires up the headline ones
  • ▸ History with ↑/↓ · clear with ⌃L
  • ▸ Free-form prompts route to the best skill in the real REPL
Want the real one? Open the Portal →
agen-it · live demo (mock)
Agen-IT REPL · demo session · v5.0 (mock)
Type /help for commands, or anything else to see how it feels.
[demo] flow>
try:

/ Commands

92+ slash commands

A REPL designed like an OS shell — every workflow, integration, and persona is one command away.

V-Model
/run/req/arch/code/test/audit/swe1/swe2/swe4/swe5
Autonomous
/goal/overnight/orchestrate/feature/squad/agent/jobs
Memory & Soul
/memory/soul/global/mempalace/graph/decay/events
Compliance
/compliance/cve/qa/testgap/traceability/sandbox
Hardware
/configure/configure-apply/logic2
Integrations
/mcp/extension/marketplace/skills/backend/federation

/ Tools

What we ship

Production tools from the Agen-IT organization — composable, profile-driven, and battle-tested in embedded workflows.

Use them in the portal →

AgenIT

GitHub ↗

Profile-driven AI development co-pilot

TypeScript orchestrator over Gemini CLI with persistent soul memory, ASPICE V-Model pipeline, Mission Squad, and a marketplace for MCP servers, skills, and extensions.

TypeScriptGemini CLIMCPASPICEInk REPL

AgenIT Arduino-as-HIL

GitHub ↗

Arduino-based Hardware-in-the-Loop testing node

Lightweight HIL implementation that turns Arduino boards into hardware-in-the-loop test nodes — pairs with AgenIT's /logic2 and JLink integration for embedded validation.

PythonHILArduinoTestingEmbedded

/ Use Cases

Where teams ship with Agen-IT

Automotive embedded

AUTOSAR / CAN / LIN drivers with ASPICE-compliant traceability from CRS to test report.

Safety-critical firmware

MISRA-aware code generation with HIL validation loops and audit-ready evidence.

Long-running engineering loops

Overnight autonomous Plan → Code → Test → Evaluate cycles with git-backed rollback.

Web & product teams

Same V-Model discipline, swapped profile: user stories, Playwright tests, OWASP/a11y audits.

/ Install

Up and running in minutes

Node ≥ 20, pnpm ≥ 9, and a Gemini CLI OAuth login — no API keys, no vendor lock-in. Python is only needed for the bundled hardware tools.

  • ▸ Node.js ≥ 20
  • ▸ pnpm ≥ 9
  • ▸ Gemini CLI ≥ 0.37 (OAuth)
  • ▸ Python 3.10+ (hardware tools, optional)
  • ▸ SEGGER J-Link (hardware debug, optional)
install.sh
# 1. Install Gemini CLI
$ npm install -g @google/gemini-cli
$ gemini auth login

# 2. Clone + build Agen-IT
$ git clone https://github.com/AgenticalXx/AgenIT
$ cd AgenIT && bash install.sh

# 3. Initialize your project
$ agenit init
$ agenit
[my-project] flow> /req requirements/CRS.pdf
[my-project] flow> /arch
[my-project] flow> /code
[my-project] flow> /test
[my-project] flow> /audit

Ready to put agents to work?

Sign in to the Agen-IT portal to access our tools, manage profiles, and run agentic workflows across your projects.

Go to Portal →Browse on GitHub