# software-mansion/argent

An agentic toolkit to control, debug, and profile iOS and Android apps. Made by Software Mansion.

Repository: https://github.com/software-mansion/argent
Canonical: https://ross.abutalabs.com/products/argent
Homepage: https://argent.swmansion.com
Language: TypeScript
License: Apache-2.0
License Family: permissive
Topics: agentic, ios, react-native, android
Last push: 2026-08-26T15:27:12+00:00

## Health v2 (maintenance only)
Score: 82/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 99, longevity 13
- inputs: {"age_days": 191, "days_push": 7, "days_rel": 8, "gap_med": 4.5, "n_releases_24m": 29}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2249, forks 86 (observed 2026-08-28T04:06:30.221612+00:00)

## What it is
Argent is an agentic toolkit by Software Mansion that gives AI assistants (via an MCP server and skills) direct control of iOS simulators, Android emulators/devices, TVs, and Electron/Chromium apps. It lets agents build, launch, interact with, debug, and profile mobile and desktop apps autonomously from the CLI.

## Use cases
- let an AI agent tap through my app on an iOS simulator and reproduce a bug
- automate UI interactions on Android emulator from an AI coding assistant
- profile React Native app performance and trace slow commits to native frames
- record and replay manual test flows deterministically
- take screenshots and diff them for visual regression
- inspect React component tree and network requests at runtime via an agent

## When to choose
- you develop iOS, Android, React Native, or TV apps and use an AI agent in your workflow
- you want autonomous bug reproduction, manual testing, or profiling without leaving your editor
- you need an MCP server connecting agents to simulators, emulators, and devices

## When to avoid
- you need CI-scale automated test suites rather than agent-driven interaction
- your targets are platforms outside iOS, Android, TV, or Chromium/Electron
- you don't use any AI agent harness or MCP-capable editor

## Facets
- artifact type: framework
- maturity: active
- function: mcp, agent-framework, testing, e2e-testing, monitoring, developer-tools, cli
- domain: mobile-development, developer-tools, testing
- platform: windows, cli
- tags: react-native, mcp-server, ios-simulator, android-emulator, profiling, ui-automation, tvos, visual-regression, software-mansion, automation, ai-agents, ios, android, macos, linux, nodejs, electron

## Member repositories
- software-mansion/argent (main) score 82

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:30.221612+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:43:54.832421+00:00, confidence not recorded.
  - readme: https://github.com/software-mansion/argent (fetched 2026-08-28T04:06:30.221612+00:00, sha e5e4259255ef)
  - homepage: https://argent.swmansion.com (fetched 2026-08-29T10:24:17.733215+00:00, sha 80f80ccb604a)
  - site_page: https://docs.swmansion.com/argent/docs/fundamentals/getting-started (fetched 2026-08-29T10:24:17.735940+00:00, sha 9024d39040f8)
- Data as of 2026-08-30T08:39:29.467469+00:00.