> ## Documentation Index > Fetch the complete documentation index at: https://docs.mobster.dev/llms.txt > Use this file to discover all available pages before exploring further. # Errors, statuses and exit codes > Every exit code, task status, check verdict and HTTP error, and what to do about each. Mobster tells you why something ended in four places: the exit code of a command, the `status` of a task, the `verdict` of a check, and the error code of an HTTP answer. [Troubleshooting](/troubleshooting) starts from what you see instead. ## Exit codes Every command uses the same small set. A script can branch on them without reading the output. | Code | `mobster run` | `mobster verify` | `mobster test` | Other commands | | - | - | - | - | - | | 0 | The task finished, or a preview was shown | `passed` | Every check passed (flaky ones too, without `--strict`) | It worked | | 1 | The task didn't finish: its `status` says why | `failed` | A check failed, or was flaky with `--strict` | A check or an action failed | | 2 | A usage error | `needs_review` | None failed or couldn't run, and one needs review | A usage error | | 3 | No phone answered at the WebDriverAgent address, or `--device` names no device that's ready | `couldnt_run`, a usage error included | None failed and one couldn't run, or a usage error | No such device, or no Xcode tools | | 130 | Stopped with `ctrl+c` | Stopped: `couldnt_run`, class `stopped` | Stopped, with the reports still written | Stopped | SIGTERM exits 143 and SIGHUP 129, the same way as `ctrl+c`. `mobster verify` and `mobster test` exit 3 on a usage error, not 2, because 2 means a check needs review. `mobster test` decides from the checks that aren't quarantined, in this order: 1 when one failed (or was flaky, with `--strict`), else 3 when one couldn't run, else 2 when one needs review, else 0. `mobster chat` adds one: 2 when the task waits for you (an approval or a question), and 4 when it's still working when `--wait` runs out. [CLI reference](/reference/cli) gives each command's codes. ## Statuses A task's result has a `status`. `mobster run --json` prints it in its last line, and the HTTP API and the Mac app's History show it too. | `status` | What happened | Exit code | | - | - | - | | `completed_unverified` | Mobster's agent judged the goal met. Nothing checked it independently | 0 | | `expected_text_visible` | The `--expected-text` appeared | 0 | | `preview` | A preview, without `--execute` | 0 | | `completion_not_confirmed`, `inconsistent_completion` | It looked done but couldn't be confirmed. `data_status` says why an answer is missing | 1 | | `blocked`, `no_progress`, `needs_clarification`, `needs_text_helper` | It couldn't go on | 1 | | `max_steps`, `timeout`, `spend_cap` | It hit a limit | 1 | | `stopped`, `error` | Stopped, or failed. The reason says whether an action may have run | 1 (130 with `ctrl+c`, 3 when no phone answered) | [Troubleshooting: tasks](/troubleshooting#tasks) has what to do for the ones people meet most. ## Verdicts A check ends with one of four verdicts. The verdict comes from your assertions on the accessibility tree, never from a model. | Verdict | Exit code | What it means | | - | - | - | | `passed` | 0 | Every assertion held on one settled read, with the app in front | | `failed` | 1 | An assertion didn't hold, or the app wasn't in front at the end | | `needs_review` | 2 | Nothing decided the run: the check has no assertions, or every one held before the steps ran | | `couldnt_run` | 3 | The check is invalid, or this Mac couldn't run it | ### Reason classes `reason.class` in the result says why a check failed, couldn't run or needs review. | `reason.class` | Verdict | What it means | | - | - | - | | `usage` | `couldnt_run` | The check or an assertion is invalid, or it has steps and no key to run them | | `model` | `couldnt_run` | No key for Mobster's agent, or the model failed or ran out of credit | | `environment` | `couldnt_run` | Xcode, the iOS runtime or the device type is missing | | `simulator` | `couldnt_run` | The simulator couldn't be created or booted | | `wda` | `couldnt_run` | WebDriverAgent couldn't be built, started or read | | `install` | `couldnt_run` | The `.app` is invalid, is a device build, or didn't install | | `launch` | `couldnt_run` | The app didn't launch | | `busy` | `couldnt_run` | Every one of Mobster's simulators is in use | | `device_lost` | `couldnt_run` | `mobster test`: a phone stopped answering while its check ran, or earlier in the run. The check isn't run again | | `budget` | `couldnt_run` | The steps hit the spend cap and an assertion fails | | `stopped` | `couldnt_run` | You stopped the run, or a key-less session sat idle for 15 minutes | | `internal` | `couldnt_run` | An unexpected error. The traceback is in `run.log` | | `app_crashed` | `failed` | The app crashed during the run. The message names the exception and the first frame in your code; the report shows the stack and the log around it | | `app_not_running` | `failed` | The app in front at the end isn't the app under test: the flow left it, or it crashed where Mobster reads no log | | `blocked` | `failed` | An assertion failed, and the steps ended blocked, out of time or out of steps | | `assertion` | `failed` | An assertion failed | | `no_assertions` | `needs_review` | `expect` is empty | | `held_before_flow` | `needs_review` | Every assertion already held right after launch | [Checks](/checks#verdicts-and-exit-codes) has the order these rules apply in, and [Troubleshooting: checks](/troubleshooting#checks) what to do about each. ## HTTP codes The local API that `mobster serve` and the Mac app run answers errors with a status and a short `code`. The codes you're most likely to meet: | Code | Status | What it means | | - | - | - | | `app_only` | 403 | The Mac app runs the server, and only the app may do this: approve, change safety settings, put a file on the phone or reset the Wi-Fi link | | `plan_required` | 403 | The Mac app, or the agent inside it, runs the server, and this Mac's account has no active plan or free trial, so no task starts. The error says what to do: open Mobster to start your free trial, or choose a plan at mobster.dev/pricing ([What needs a plan](/account#what-needs-a-plan)) | | `needs_cable` | 409 | The phone is on Wi-Fi, and this needs the cable. Plug the iPhone in and try again | | `no_file_sharing` | 409 | That app has no folder in Files on the iPhone | | `no_device` | 409 | No iPhone is connected | | `locked` | 409 | The iPhone is locked. Unlock it, then try again | | `device_busy` | 409 | A task is running on that device | | `approval_pending` | 409 | The task waits for an approval. Answer it first | | `not_found` | 404 | No such item | | `tools` | 503 | The iPhone tools are missing on this Mac | | `unavailable` | 503 | Mobster couldn't open its data | Each area lists its own codes beside its routes: [Scripts and the API](/scripting#the-http-api), [Conversations](/scripting#conversations), [Memory](/memory) and [Files](/files). This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.