# iOSWorld

Native iPhone-simulator benchmark with shared user data, application workflows, agent runners, and rubric-based scoring.

- Catalog URL: https://awesomecua.github.io/resources/github-ljang0-iosworld/
- Original source: https://github.com/ljang0/iOSWorld
- Kind: benchmark
- Review: curated
- Scope: core
- Environments: ios
- Topics: evaluation, cross-app, memory
- Evidence checked: 2026-10-16T10:42:12Z
- ID: github:ljang0/iosworld

## Limitations

Inclusion does not establish reproducibility or endorsement.

- Record JSON: https://awesomecua.github.io/resources/github-ljang0-iosworld/record.json

## Structured profile

- actions: touch actions, MCP tools (optional)
- evaluation: Rubric satisfaction per task; full-task pass requires every criterion to be satisfied.
- focus: personalized phone tasks, cross-app workflows
- observations: screenshots, accessibility XML (optional)
- setting: interactive

### Count observations

- 133 tasks — tasks.json benchmark (reported; source: https://raw.githubusercontent.com/ljang0/iOSWorld/HEAD/README.md)
- 26 applications — bundled SwiftUI apps (reported; source: https://raw.githubusercontent.com/ljang0/iOSWorld/HEAD/README.md)

Vision-only, vision-plus-XML, and MCP modes expose different information and tools. The documented default evaluator is model-based; compare the same mode and judge.

## License declarations

- code: Apache-2.0 — https://raw.githubusercontent.com/ljang0/iOSWorld/HEAD/LICENSE
- data: not recorded
- document: not recorded
- weights: not recorded

## Primary evidence

- https://raw.githubusercontent.com/ljang0/iOSWorld/HEAD/README.md (retrieved 2026-10-16T10:42:12Z; SHA-256 706f9d770b65a1238e56f74f7dd5089c7756cc88e0ac09d80c7b2b3085a09b27)
- https://raw.githubusercontent.com/ljang0/iOSWorld/HEAD/LICENSE (retrieved 2026-10-16T10:48:41Z; SHA-256 fa9fe1e31ed8da21d0b51621d3a3669ced4f0e42353aaf4bd8be1615d200eaac)