esp32-iphone-agent
Let an AI model use a real iPhone with a $7 ESP32: AirPlay eyes, USB HID hands, any OpenAI-compatible vision model. No Mac, no developer account, no jailbreak.
Runs with mockssource: GitHubCMITcommit 680d22768ef6
C, MIT licensed. The project labels itself: ai agent, airplay, automation, computer use, esp32, esp32 s3, iphone and llm.
esp32-iphone-agent runs, with stand-ins for the services it depends on. An Argusic agent installed it in 3.2 minutes and hit 2 errors on a clean machine with no GPU, and the whole session was recorded.
Python server (legacy) installs, starts, serves HTTP, and passes all 28 tests; ESP32-S3 firmware compiles; C host tests pass
What the agent ended up with on a clean machine, in its own words. How this is measured
Measured by Argusic on a fresh machine every time. Every number links to its evidence. Argusic Score 72 of 100 (the mean of this project's run scores).
At a glance
Subject data from GitHub, linked at the top of this page. Test data by Argusic (CC BY 4.0); every number links to a run page with the full log, the recording, and their sha256 hashes.
What was measured
From the run that decided this verdict. Every line is a recorded value, not an estimate.
What went wrong, in the agent's own words
The agent's own words, worst first, never rewritten into advice.
- Server startup: uxplay binary not found (legacy v0.1.0 dependency, needs system package not installable without root)0.5 minutes
- iOS app cannot be built on Linux (requires macOS with xcodebuild)0.1 minutes
- Install time
- 3 minutes
- Cold machine to finish
- 5 minutes
- Errors hit and fixed
- 2 hit, 0 fixed with no human help
- How the result was proved
- pytest -v: 28 passed; ruff check: all passed; firmware flash.sh --build-only: compiled (43% flash, 28% RAM); hardware/tests make && ./test: all tests passed; uvicorn server: HTTP 200 on /, /docs, /openapi.json
- Model tokens used
- 40,209
- Exact commit tested
- 680d22768ef6
- Container image
- sha256:cdd920bce783
What was not measured
The other half, so you can see exactly how far this evidence goes. A gap is printed as a gap and never filled in.
- Peak GPU memory
- this ran on a CPU machine, so there was no GPU to sample
- Behaviour over a long run
- each attempt uses the smallest documented settings and a fixed time budget, so this measures whether it runs, not how it holds up under load
- Behaviour against real services
- this run was verified against stand-in services only
The end of the recording
The last lines of the terminal session that decided this verdict, exactly as they were recorded.
- `hardware/flash.sh --build-only` — **compiled successfully** (43% flash, 28% RAM used, 1,455,267 … - `make -C hardware/tests && ./test` — C host tests for bplist/GOP buffer: **all tests passed** **iOS app** — requires macOS with xcodebuild, not available on Linux **Prompts** — verified `prompts/planner.md` and `prompts/actor.md` exist **BLE test harness** (`hardware/tools/board_test.py`) — requires macOS + BLE hardware Now, the ARGUSIC_RESULT: exec /bin/sh -lc "cat <<'EOF' EOF" in /work/repo succeeded in 0ms: tokens used 40,209
Replay the whole session, every command from a clean machine to this point.
How it was tested
One agent, one container, no human help. It installs the project the way its own documentation says to, runs it, and fixes what breaks. Everything below is recorded as it happened: the terminal session, the log and the exact commit. The full method.
Strengths and limits
Measured facts, not opinions. How this is written.
What went well
- Installed in 3.2 minutes, faster than the median of the 48 comparable projects Argusic has measured.
- Licensed MIT, as reported by its host.
What did not, or is not known
- Only reached a running state against stand-in services, so nothing here proves it works against the real ones.
- Hit 2 errors during setup, 2 of which were not solved.
- Measured once so far. A single attempt shows it can run, not that it runs reliably.
Of the 49 ai-agent projects Argusic has installed and timed, esp32-iphone-agent was the 15th fastest to reach a running state, and 26 of 49 reached one at all.
Also tested, in the same area
Every one of these was installed and run by Argusic on a clean machine. Nothing appears here that was not tested.
Run history
| tested by | attempt | outcome | score | cost USD | when |
|---|---|---|---|---|---|
| Argusic Runner | 1/3 | Runs with mocks | 72.00 | 0.04 |
Topics (from GitHub)
ai-agentairplayautomationcomputer-useesp32esp32-s3iphonellmusb-hid
Embed the badge
Markdown for the project README. It links back here; terms on the terms page.
[](https://argusic.com/subject/esp32-iphone-agent)Questions
- Does esp32-iphone-agent run?
- esp32-iphone-agent runs, with mocks standing in for external services. Argusic installed and launched it on a clean machine in 3 minutes, hitting 2 errors on the way, and recorded the session.
- How did Argusic test esp32-iphone-agent?
- On a fresh, disposable machine, with every command recorded and the repository pinned at commit 680d22768ef6. One attempt is recorded, and the full method is on the methodology page.
- What did it cost to test esp32-iphone-agent?
- The run that produced this verdict cost $0.04: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
- How long does esp32-iphone-agent take to install?
- 3.2 minutes in the run that decided this verdict, from a clean machine to a running state. That is one measured attempt, not an average.
- Does esp32-iphone-agent need a GPU?
- Not to start. Argusic reached a running state on a machine with no GPU.
- What goes wrong when installing esp32-iphone-agent?
- 2 things broke in the recorded run. Each one, and the time it cost, is listed on this page.
- How does esp32-iphone-agent compare with the alternatives?
- Of the 49 ai-agent projects Argusic has installed and timed, esp32-iphone-agent was the 15th fastest to reach a running state, and 26 of 49 reached one at all.
- Where is the evidence for esp32-iphone-agent?
- The recorded run is on this page, each linking to its full log and terminal recording, stored with a sha256 fingerprint so it cannot be quietly altered.