Are you an LLM? Read llms.txt for a summary of the docs, or llms-full.txt for the full context.
Skip to content

Amodal

Agent Runtime Platform

Build agents from your repo

What is Amodal?

Amodal is an open-source agent runtime configured by a git repo of markdown and JSON. Rather than building agent infrastructure from scratch, SaaS products connect their APIs, add domain knowledge, and define skills — and the platform provides the reasoning engine, context management, and learning infrastructure underneath. The core loop is explore → plan → execute: the agent gathers data, reasons about it, and takes action, compounding knowledge with every interaction. You write configuration files. Amodal turns them into a production-grade AI agent.

How it works

An Amodal project is a folder of plain files that describe what your agent knows, what it can access, and how it should think:

  • Connections define API access. Each connection is a directory with a spec, auth config, and natural-language documentation that teaches the agent what endpoints exist and when to use them.
  • Skills are markdown files that encode expert reasoning — investigation methodologies, triage procedures, escalation logic. They activate automatically based on the user's question.
  • Knowledge is persistent domain context: environment details, team structures, historical patterns, baseline metrics. The agent loads it on demand and proposes updates as it learns.
  • Tools are custom actions beyond the built-in set — HTTP calls, multi-system chains, or arbitrary function handlers with confirmation rules and rate limits.

When you run amodal dev, the runtime compiles all of these into a system prompt and drives a state-machine agent loop against any supported LLM (Anthropic, OpenAI, Google, DeepSeek, Groq, Mistral, xAI). The agent reasons about user questions, dispatches sub-agents for data gathering (keeping its own context clean), and returns structured answers. Every interaction can feed knowledge back into the system, so the agent gets smarter over time.

npx amodal init                          # scaffold a project
amodal pkg connect slack                 # add Slack API access
echo "# Triage\n..." > skills/triage.md  # define a reasoning skill
amodal dev                               # start the runtime
amodal chat                              # talk to your agent

Build

  • Quick Start — Initialize a project, connect an API, and chat with your agent in 5 minutes.
  • CLI Reference — init, dev, chat, eval, pkg sync, and 30+ more commands.
  • Connections — 20+ pre-built plugins for Slack, GitHub, Stripe, Datadog, Jira, and more.
  • Skills — Author expert reasoning frameworks in Markdown.
  • Tools — Define custom HTTP, chain, and function tools.
  • Knowledge — Persistent domain context that grows with use.

Run & Embed

  • Providers — Anthropic, OpenAI, Google Gemini with automatic failover.
  • React SDK — Embed agents in your product with @amodalai/react.
  • Architecture — The explore-plan-execute loop, context isolation, and multi-agent dispatch.