# Swipium (npm · swipium)

Mobile QA on local Android Emulators and iOS Simulators: autopilot tests, evidence, reports.

- Trust score: 90/100 (high trust)
- Change this week: +4
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-04

## Components

- npm · `swipium`: 90/100 (this document), [markdown](https://verifymcp.io/servers/geropalombo-swipium/swipium.md), [page](https://verifymcp.io/servers/geropalombo-swipium/swipium)

## Channel facts

- Registry: `npm`
- Package: `swipium`
- Version: `2.4.0`
- Transport: `stdio`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-11-04.

- **Supply Chain Security**: 100/100
  - No malware found by supply-chain analysis.
  - No known CVEs affecting this package version or its production dependencies.
  - No install/post-install scripts declared.
  - 3 of 23 dependencies flagged as unhealthy.
- **Provenance & Transparency**: 100/100
  - Source repository is publicly reachable at the declared URL.
  - Cryptographically verified build provenance (signed, bound to GeroPalombo/swipium).
  - Clear OSI-approved license (MIT).
  - Actively maintained (last published 2 days ago).
  - Publishes a security disclosure policy (SECURITY.md).
- **Schema Quality & AI Usability**: 71/100
  - 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name).
  - AI-judged instruction clarity (good).
  - Context-footprint check failed: tool/resource definitions use about 9848 tokens (~175/item across 56 items; 55 tools + 1 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 69/100
  - Stability check failed: the tool surface changed between 2.3.0 and 2.4.0: 0 tool removals, 1 breaking changes, 0 additions.
- **Tool Coverage**: 89/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 61% of tool parameters carry a description.
  - Structured output schemas are declared (7% of tools); any adoption earns full credit.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - We read all 55 captured tool definition(s), and no name or description among them implies an irreversible operation.
  - An AI judge read all 57 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a current MCP spec version (2026-07-28).

## Install

### How do I install the Swipium MCP server?

Swipium runs locally as an npm package, launched with npx -y swipium. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add geropalombo-swipium -- npx -y swipium
```

### Cursor

```json
{
  "mcpServers": {
    "geropalombo-swipium": {
      "command": "npx",
      "args": [
        "-y",
        "swipium"
      ]
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "geropalombo-swipium": {
      "command": "npx",
      "args": [
        "-y",
        "swipium"
      ]
    }
  }
}
```

### Codex

```bash
codex mcp add geropalombo-swipium -- npx -y swipium
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "geropalombo-swipium": {
      "type": "local",
      "command": [
        "npx",
        "-y",
        "swipium"
      ],
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add geropalombo-swipium --command npx --arg -y --arg swipium
```

### Hermes

```yaml
mcp_servers:
  geropalombo-swipium:
    command: "npx"
    args: ["-y", "swipium"]
```

### Netclaw

```json
{
  "McpServers": {
    "geropalombo-swipium": {
      "Transport": "stdio",
      "Command": "npx",
      "Arguments": [
        "-y",
        "swipium"
      ]
    }
  }
}
```

### Vellum

```bash
assistant mcp add geropalombo-swipium -t stdio -c npx -a -y swipium
```

### Other

```json
{
  "mcpServers": {
    "geropalombo-swipium": {
      "command": "npx",
      "args": [
        "-y",
        "swipium"
      ]
    }
  }
}
```

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-11-04 (score 90, +1)

No change was recorded against any check on this day. Stability & Change Management went from 66 to 69.

### 2026-11-03 (score 89, +15)

- [security improvement] Malware scan: unverified → pass

### 2026-11-02 (score 74, −14)

- [security regression] Malware scan: pass → unverified
- [functional] Package version: 2.3.5 → 2.4.0

### 2026-10-31 (score 88, +1)

No change was recorded against any check on this day. Stability & Change Management went from 52 to 56.

### 2026-10-30 (score 87, +1)

- [functional improvement] Security disclosure: unverified → pass

### 2026-10-29 (score 86, 0)

- [functional regression] Security disclosure: pass → unverified

### 2026-10-27 (score 86, +1)

No change was recorded against any check on this day. Stability & Change Management went from 39 to 42.

### 2026-10-25 (score 85, +1)

No change was recorded against any check on this day. Stability & Change Management went from 32 to 36.

## MCP tools (55)

### `qa_act` (~392 tokens)

Act on the screen

Tap, type, clear, swipe, scroll, press a key, open a URL or wait: one UI step, then it waits for the screen to settle and reports changed/settled, snapshot quality, health, new elements. Target: an @eN ref from qa_snapshot, text, id, a native selector (WDA iOS), or x/y. For 3+ known steps, batch them in one qa_flow_run {flowYaml}. Details: docs/tools.md#qa_act.

Input parameters:

- `action` (string, required)
- `direction` (string): swipe/scroll (required)
- `durationMs` (number): tap: press duration (coordinate taps default ~100ms).
- `for` (object): wait: {settled:true} (default), or until an element (ref/text/id/selector) is visible; gone:true waits until it is not.
- `ignoreOverlay` (boolean): tap: tap even if an overlay/keyboard covers the target.
- `key` (string): press (required)
- `maxScrolls` (number): scroll: max swipes (default 8).
- `mode` (string): type: replace (default) clears first.
- `observe` (string): diff (default after a snapshot): elements added/removed; full: capped list; none: verdicts only.
- `sessionId` (string, required)
- `submit` (boolean): type: press enter after.
- `target` (object): tap/type/clear (required); swipe: optional start point.
- `text` (string): type (required)
- `timeoutMs` (number): wait: max ms (default 8000, max 50000; larger values are clamped); others: settle-wait cap.
- `untilVisible` (object): scroll: stop once visible (checked before the first swipe).
- `url` (string): open_url (required)

Output parameters:

- `action` (string)
- `changed` (boolean)
- `elements` (array)
- `ok` (boolean)
- `settled` (boolean)

### `qa_app_control` (~115 tokens)

App lifecycle control

Control the app under test: launch, foreground, background, force_stop, restart (force_stop + launch, for persistence checks), clear_data and fresh_start (wipe data: destructive, consent-gated; RN/Expo also need acknowledgeBundleRisk).

Input parameters:

- `acknowledgeBundleRisk` (boolean): Accept that wiping an RN/Expo debug build may leave it unable to load its bundle.
- `action` (string, required)
- `approve` (boolean)
- `consentId` (string)
- `sessionId` (string, required)

### `qa_app_map_build` (~134 tokens)

Build / update the app knowledge map

Build or update .swipium/app-map.json: a static scan of routes and screens (Expo Router, React Navigation, Android, SwiftUI/UIKit, Flutter), plus the latest exploration graph when sessionId is given.

Input parameters:

- `forceRescan` (boolean): Re-scan even if the map is current.
- `includeCodeIndex` (boolean): Persist a code symbol index (default true).
- `mode` (string): static_only | runtime_merge | full (default).
- `projectRoot` (string): Default: the session or resolved project root.
- `sessionId` (string): Adds its latest exploration graph.

### `qa_app_map_feature_scope` (~151 tokens)

Scope testing to a feature

Resolve a feature (featureId, or a query like "checkout") to a test scope: code, screens, existing tests, coverage gaps, strategy, ranked candidates. Works without a map (code scan). Read-only.

Input parameters:

- `featureId` (string): Exact map feature id (or use query).
- `includeCode` (boolean): query: scan source code (default true).
- `limit` (number): Max items per list (default 8).
- `platform` (string)
- `projectRoot` (string): Project root when no session exists.
- `query` (string): e.g. "checkout flow".
- `sessionId` (string): Adds runtime screen-graph evidence.

### `qa_app_map_query` (~78 tokens)

Query the app knowledge map

Search the app map (features, screens, code, tests) with a natural-language query. Returns ranked results with sources and the suggested next tool call.

Input parameters:

- `intent` (string)
- `limit` (number)
- `projectRoot` (string)
- `query` (string, required)
- `sessionId` (string)

### `qa_app_map_read` (~93 tokens)

Read the app knowledge map

Read a compact app-map section: summary (default) | screens | features | auth | automation | testSuite | full; featureId or screenId drills into one node. Large sections come back as a resource URI.

Input parameters:

- `featureId` (string)
- `projectRoot` (string)
- `screenId` (string)
- `section` (string)
- `sessionId` (string)

### `qa_app_map_update` (~127 tokens)

Update the app knowledge map

Targeted app-map edits without a rebuild: add a note, register test cases, link an automation suite, set the environment, or override a feature's coverage. A matching id/path is overwritten.

Input parameters:

- `automationSuite` (object)
- `environment` (string): App environment, e.g. 'test' | 'staging'.
- `featureCoverage` (object)
- `note` (string): A free-text user note added with user_note provenance.
- `projectRoot` (string)
- `sessionId` (string)
- `testCases` (array)

### `qa_build` (~176 tokens)

Plan or run a build from source

Build the app from source. mode:"plan" (default, no side effects): commands, cwd, expected artifacts, toolchain status. mode:"run" (consent-gated job, needs sessionId): runs them, stores the build log, re-resolves the artifact. Failures are typed (GRADLE_FAILED, XCODEBUILD_FAILED, ...) and are not test failures.

Input parameters:

- `approve` (boolean): run
- `consentId` (string): run
- `mode` (string)
- `platform` (string, required)
- `projectRoot` (string): plan: resolve the project without a session.
- `sessionId` (string): Required for run (stores the build log).
- `timeoutMs` (number): run: per-step timeout (default 1200000).
- `variant` (string)

### `qa_bundletool` (~182 tokens)

Convert .aab > installable .apk

Convert an .aab into an installable APK with bundletool (cached in .swipium/artifacts/) before qa_prepare_target. Default: a universal APK; connectedDevice:true builds a device APK set, install:true installs it (consent-gated). Runs as a job.

Input parameters:

- `aab` (string): Default: the best .aab in the project.
- `approve` (boolean)
- `connectedDevice` (boolean): Device-specific APK set (--connected-device).
- `consentId` (string)
- `device` (string): adb serial (default: session device, else SWIPIUM_DEVICE, else the only one).
- `force` (boolean): Rebuild even if cached.
- `install` (boolean): With connectedDevice: install-apks on the device (consent-gated).
- `sessionId` (string, required)

### `qa_check_health` (~55 tokens)

Check app health

Health check of the current screen: crash, ANR, error boundary, error surface, app still in the foreground. High-severity findings are real bugs. qa_act runs it automatically.

Input parameters:

- `sessionId` (string, required)

### `qa_clear_overlay` (~88 tokens)

Clear an overlay

Clear what blocks the screen (keyboard, LogBox, permission dialog, toast); strategy "auto" handles the topmost. targetRef checks whether that element is still obstructed. Returns what was cleared and what remains.

Input parameters:

- `sessionId` (string, required)
- `strategy` (string): default "auto"
- `targetRef` (string): @eN to check obstruction before/after

### `qa_continue_from_blocker` (~165 tokens)

Resume after providing input

Answer a needs_input question (credentials, OTP, platform, device, monorepo target) and get the next call. Secrets (passwords/OTP/tokens) are redacted at once and never echoed or logged (flows use ${VAR}). Non-secret choices map onto the re-invocation args. Returns accepted, ignored[], projectRoot, nextAction; with no new consent needed it starts the resumed job (jobId).

Input parameters:

- `kind` (string, required): The needs_input kind (e.g. credentials, monorepo_target).
- `secretFields` (array): Secret keys (default: password/otp/token-like names).
- `sessionId` (string, required)
- `values` (object): Field name > value. Secret fields are redacted on receipt.

### `qa_device_info` (~189 tokens)

Device info

Read-only device info. Android: model, SDK, ABIs, locale, screen, orientation (listPackages:true adds third-party packages; logs:{…} captures filtered logcat). iOS Simulator: name, runtime, state, screen.

Input parameters:

- `listPackages` (boolean): Include installed third-party package names.
- `logs` (object): Android: save logcat to an artifact instead. Scans the last `lines` (1000) lines overall; level is the minimum; package also matches a crashed app (its `Process:` PID); grep is a regex, context = lin…
- `packageFilter` (string): Substring filter for the package list.
- `sessionId` (string, required)

### `qa_doctor` (~137 tokens)

QA environment doctor

Diagnose the local QA toolchain (Node, Android SDK/emulator, Xcode/simctl, WDA) and stale-client symptoms; run it first when setup fails.

Input parameters:

- `client` (string): Tailor setup hints to this MCP client.
- `expectedSchemaHash` (string): Surface hash you expect; a mismatch means a stale client.
- `expectedToolCount` (number): Tool count you expect; a mismatch means a stale client.
- `expectedVersion` (string): Version you expect; a mismatch means a stale client.
- `platform` (string): Default both on macOS (ready if either is), else android.

### `qa_explain_blocker` (~98 tokens)

Explain a blocker

Explain a failureCode in plain language: what it means, who owns the fix (app / environment / Swipium / user), whether it is retry-safe or Swipium can fix it, and the recovery step.

Input parameters:

- `context` (string)
- `failureCode` (string, required): failureCode from a Swipium error.
- `sessionId` (string): The blocked session (qa_status then moves past the blocker).

### `qa_explore` (~378 tokens)

Guided exploration

Explore the prepared app automatically as a job (poll qa_job_status): taps ranked safe actions, checks health, builds a screen graph, updates the app map, and records taps for qa_generate. Destructive actions (delete/pay/send/logout) are skipped unless one candidate is approved; an auth wall without credentials returns needs_input.

Input parameters:

- `accountCycle` (boolean): Allow logout on a disposable generated account (needs allowGeneratedData).
- `allowGeneratedData` (boolean): Allow generated test data (test/staging).
- `approve` (boolean)
- `confirmHighImpact` (boolean): Required for payment/send/permission/delete candidates.
- `consentId` (string)
- `depth` (integer): Default 3. Not a depth limit: 1 turns off going back from a dead-end screen.
- `destructiveCandidate` (object): approved_destructive_candidate: the candidate from a dry_run_destructive run.
- `generateSuite` (boolean): Also write + compile a POM suite from promoted paths.
- `goal` (string): Focus, e.g. "exercise the main tabs".
- `includeTextEntry` (boolean): Type into fields that have a value source (default false).
- `maxActions` (integer): Default 20.
- `maxDurationMs` (integer): Default 360000.
- `maxScreens` (integer): Default 12.
- `safeMode` (string): strict (default) | balanced | dry_run_destructive (list candidates, no taps) | approved_destructive_candidate (one candidate, consent-gated). approved_destructive is refused.
- `sessionId` (string, required)
- `stopOnAuth` (boolean): needs_input at an auth wall without credentials (default true).
- `strategy` (string): crawl (default) | task_planner (infer QA tasks first) | hybrid.

### `qa_first_run` (~193 tokens)

First run: plan or continue through login/onboarding

Get past a first-run gate (login, sign-up, OTP, onboarding, permissions, paywall). mode:"plan" (default, read-only) classifies the screen and returns a safe plan. mode:"continue" runs bounded steps: generated data only in test/staging, never purchases, one needs_input question on OTP. Needs a prepared device.

Input parameters:

- `allowGeneratedAccount` (boolean): Override the policy on creating a throwaway account.
- `maxDurationMs` (number): continue: time cap in ms.
- `maxSteps` (number): continue: step cap.
- `mode` (string)
- `sessionId` (string, required)
- `testDataPolicyPath` (string): Test-data policy JSON (default .swipium/test-data-policy.json).
- `until` (string): continue: one_step (default) | until_gate (paywall/OTP/permission) | until_home.

### `qa_flow_check` (~140 tokens)

Check a flow

Validate a flow (flow name/path or flowYaml) without running it: schema errors with the offending step, plus warnings. Per-backend preview: qa_flow_run mode:"plan".

Input parameters:

- `ci` (boolean): Add CI preflight warnings (missing variables, forbidden mutating steps).
- `flow` (string): Flow name under .swipium/flows, or a path to a .yaml file.
- `flowYaml` (string): Inline flow YAML (instead of a file).
- `platform` (string): Authoring target for platform-aware warnings.
- `projectRoot` (string): Absolute app root.
- `sessionId` (string)

### `qa_flow_compile` (~140 tokens)

Compile a POM suite to runnable flows

Compile a POM suite on disk (.swipium/suites/<suite>.yaml) into runnable flows under .swipium/flows/ and validate them. No session needed; qa_generate target:"suite" already compiles what it generates.

Input parameters:

- `force` (boolean): Overwrite .swipium/flows/<slug>.yaml even when it was edited by hand since the last compile (default: keep it, write <slug>.yaml.new).
- `projectRoot` (string)
- `sessionId` (string)
- `suite` (string): Suite file relative to .swipium/ (default suites/smoke.yaml).

### `qa_flow_repair` (~125 tokens)

Repair a failed flow step

Repair a failed flow step: suggest a stronger locator and app changes (testID/accessibilityIdentifier); apply:true patches simple selector steps at high/medium confidence (low confidence: refused, applied:false + a note).

Input parameters:

- `apply` (boolean): Patch the flow file when safe (default false).
- `failedStep` (integer, required): Zero-based failed step index from qa_flow_run.failedAtStep.
- `flow` (string): Flow name/path under .swipium/flows.
- `flowYaml` (string): Inline flow YAML.
- `sessionId` (string, required)

### `qa_flow_run` (~268 tokens)

Run a flow (or preview its execution plan)

Run a YAML flow (saved, or inline flowYaml to batch known steps) on the prepared app; mode:"plan" previews per-backend support without a device. Server-side, fail-fast, with setup/teardown; mutating and OCR steps are consent-gated; a failure returns the step, screenshot, failureCode, and health. Pass flow (name/path) or flowYaml, plus variables. Steps: docs/flows.md#step-reference.

Input parameters:

- `appium` (object): plan: Appium session hints.
- `approve` (boolean)
- `backend` (string): plan: only this backend (default all five).
- `consentId` (string): run: consent for mutating / OCR steps.
- `flow` (string): Flow name under .swipium/flows, or a .yaml path.
- `flowYaml` (string): Inline flow YAML.
- `mode` (string)
- `projectRoot` (string): plan: absolute app root for flow names.
- `repeat` (integer): run: repeat N times to classify flakes.
- `sessionId` (string): Required for run.
- `variables` (object): run: ${VAR} values (over session inputs and SWIPIUM_* env); secret-like names are redacted.

### `qa_generate` (~433 tokens)

Generate test assets from recorded actions

Generate reusable test assets from the actions recorded in this session. target: flow (Flow V2 YAML), pom (page objects), suite (POM suite under .swipium/, compiled to flows, replay-gated), testcases (TC catalog), appium (WebdriverIO or Python project; can bootstrap from projectRoot). mode:"plan" previews. Durable repo suite: qa_suite_generate.

Input parameters:

- `approve` (boolean)
- `backend` (string): appium plan: preferred backend.
- `bootstrap`: appium: record actions first when none exist.
- `brittleThreshold` (number): appium: max brittle-locator % (default 40).
- `budgetProfile` (string): flow
- `candidateOnly` (boolean): appium: brittle locators do not fail validation.
- `compile` (boolean): suite: compile to runnable Flow V2 (default true).
- `consentId` (string): suite fresh_state / appium project write.
- `device` (string): appium: device to bootstrap on.
- `feature` (string): appium: focus for plan/bootstrap.
- `format` (string): testcases (default both)
- `includeCi` (boolean): appium: also emit ci.example.yml.
- `integrateIntoProject` (boolean): appium: write into the project (consent-gated, never overwrites).
- `language` (string): appium (default auto-detected)
- `mode` (string): generate (default) or plan (read-only preview).
- `name` (string): Asset name (default from the app id).
- `platform` (string): appium
- `projectRoot` (string): appium: plan/bootstrap without a session.
- `replay` (string): suite: replay gate (default dry_run; fresh_state needs stateProfile).
- `save` (boolean): Write files (default: suite/appium true, others false).
- `sessionId` (string): Required except for target:"appium".
- `stateProfile` (string): suite: for replay:"fresh_state".
- `target` (string, required)

### `qa_geolocation` (~74 tokens)

Set location

Spoof the GPS location (Android Emulator only; consent-gated, logged as an environment change).

Input parameters:

- `approve` (boolean)
- `consentId` (string)
- `lat` (number, required): Latitude in decimal degrees.
- `lng` (number, required): Longitude in decimal degrees.
- `sessionId` (string, required)

### `qa_get_artifact` (~109 tokens)

Get an artifact

Fetch a session artifact (screenshot, report, dump, log, recording) by its swipium://session/<id>/<kind>/<name> URI when the client lacks MCP resources. Text is inline (over 1 MB: the head, or the tail for logs). Binaries return metadata; mode:"inline" returns the bytes (images as JPEG, long edge 1568 px, scale in the text).

Input parameters:

- `mode` (string)
- `uri` (string, required)

### `qa_inspect` (~59 tokens)

Inspect one element

Inspect one element: full attributes (class, ids, text, bounds, flags) of an @eN ref from the last qa_snapshot.

Input parameters:

- `ref` (string, required): e.g. "@e3"
- `sessionId` (string, required)

### `qa_ios` (~234 tokens)

iOS simulator control

Control an iOS Simulator (macOS): list, boot (binds it to the session), install (.app), launch, terminate, openurl, logs, privacy_reset, erase (wipes it). install, privacy_reset and erase are consent-gated. A slow cold boot returns status:"booting" (poll qa_wait for:"simulator_booted"). Structured iOS automation: qa_wda.

Input parameters:

- `action` (string, required)
- `app` (string): install: .app path (absolute or project-relative).
- `approve` (boolean)
- `bundleId` (string): launch/terminate/privacy_reset: bundle id.
- `consentId` (string)
- `device` (string): boot/erase: simulator UDID or exact name (other actions: the bound one only).
- `last` (string): logs: time range, e.g. 30m (default 5m).
- `service` (string): privacy_reset: service, e.g. location, photos, all.
- `sessionId` (string, required)
- `url` (string): openurl: deep link.

### `qa_issue_log` (~438 tokens)

Project issue ledger (list + lifecycle)

Track bugs across runs in the project issue ledger (.swipium/issues-log.jsonl). mode: history (default, filterable), log (record an observation; re-observing a fixed issue reopens it), mark_fixed, verify_fixed (needs current-run evidence), suppress (expires; unsuppress:true lifts it), metrics (trends). Transitions take issueId or fingerprint.

Input parameters:

- `auditCheckId` (string): verify_fixed evidence
- `category` (string): history filter; log: category hint.
- `evidenceUris` (array): log/verify_fixed: swipium:// evidence URIs.
- `failureCode` (string): log: typed code, e.g. REDBOX, ANR.
- `fingerprint` (string): Alternative transition key.
- `fixedBy` (string): mark_fixed
- `fixedInCommit` (string): mark_fixed
- `fixedInVersion` (string): mark_fixed
- `groupBy` (string): metrics bucketing (default week).
- `howFixed` (string): mark_fixed (quoted on recurrence)
- `includeSuppressed` (boolean): history/metrics: include suppressed issues.
- `issueId` (string): Transition key (from history).
- `mode` (string)
- `platform` (string): history filter; log: platform observed on.
- `projectRoot` (string)
- `reportUri` (string): verify_fixed evidence
- `sessionId` (string)
- `severity` (string): history filter; log: severity hint.
- `since` (string): history/metrics: ISO start.
- `state` (string): history filter.
- `summary` (string): log: one-line summary.
- `suppressedUntil` (string): suppress: ISO expiry.
- `suppressionReason` (string): suppress
- `suppressionScope` (string): suppress
- `testCaseId` (string): verify_fixed evidence
- `title` (string): log (required): what was observed.
- `unsuppress` (boolean): suppress: lift an existing suppression now.
- `until` (string): metrics: ISO end; suppress: = suppressedUntil.

### `qa_job_cancel` (~69 tokens)

Cancel a job

Cancel a running background job and abort its child processes (build/boot/install/record). Returns {jobId, cancelled}; cancelled:false means it already finished or is unknown. Side effects already applied are not rolled back.

Input parameters:

- `jobId` (string, required)
- `sessionId` (string, required)

### `qa_job_status` (~150 tokens)

Job status (optionally wait)

Poll or wait on a background job (qa_test_this, qa_prepare_target, qa_explore, qa_build run, ...): status, progressDetail, and result/error when finished (qa_test_this: state completed/blocked/unsafe/needs_input, reportUri, suite, smoke, health). waitMs long-polls until the job leaves "running" (else waited.timedOut:true).

Input parameters:

- `jobId` (string, required)
- `sessionId` (string, required)
- `waitMs` (integer): Block up to this many ms for the job to finish (use 45000; capped at SWIPIUM_MAX_WAIT_MS, default 50000). Default 0 = return immediately.

Output parameters:

- `artifactUris` (array)
- `error`
- `jobId` (string)
- `kind` (string)
- `ok` (boolean)
- `progressDetail`
- `result`
- `status` (string)
- `waited` (object)

### `qa_metro` (~88 tokens)

Metro dev server (debug RN/Expo)

Manage the Metro bundler for debug RN/Expo builds: status, diagnose (adds RedBox/logcat evidence and fixes), start (consent-gated: adb reverse + Metro), stop. After start, relaunch with qa_prepare_target.

Input parameters:

- `action` (string, required)
- `approve` (boolean)
- `consentId` (string)
- `sessionId` (string, required)

### `qa_mobile_audit` (~201 tokens)

Mobile QA release audit

Release-readiness audit: plan (default, no device) or execute. profile: smoke, account_cycle (create/logout/login/reset on a disposable account), store_compliance, resilience (offline/relaunch/rotation; airplane mode is consent-gated), release_gate (all). execute needs a prepared session, logs failed/blocked checks to the issue ledger, passes nothing without evidence.

Input parameters:

- `allowGeneratedData` (boolean): Required for account_cycle.
- `allowTestAccountDeletion` (boolean): Permit deleting disposable test accounts (never a real account).
- `approve` (boolean)
- `consentId` (string): execute: consent for the airplane-mode toggle.
- `mode` (string)
- `offlineMode` (boolean): resilience: drive offline state.
- `profile` (string, required)
- `projectRoot` (string)
- `sessionId` (string)
- `sourceRevision` (object)
- `targetApp` (string)

### `qa_network` (~84 tokens)

Network state control

Test offline/online behavior via airplane mode (Android 11+). action: status, offline/online (consent-gated), restore. The original state is restored on qa_report, on restore, and on server shutdown.

Input parameters:

- `action` (string, required)
- `approve` (boolean)
- `consentId` (string)
- `sessionId` (string, required)

### `qa_note` (~192 tokens)

Record a test outcome

Record the outcome of one workflow for the report. Use outcome:"blocked" + missingPrecondition rather than a false failure; category says why. A pass/fail note captures a checkpoint screenshot and links the steps since the last note; extra evidence goes in artifactUris (this session's URIs only, others are dropped).

Input parameters:

- `artifactUris` (array)
- `category` (string): Default for outcome:"fail" is app_bug; use mcp_limitation for tool problems.
- `missingPrecondition` (string): e.g. "no saved flight exists"
- `outcome` (string, required)
- `reason` (string)
- `recommendedSetup` (string)
- `requiredState` (string)
- `sessionId` (string, required)
- `verifiedVisually` (boolean): The pass was confirmed from an attached screenshot.
- `workflow` (string, required): e.g. "Delete saved flight"

### `qa_orientation` (~43 tokens)

Set orientation

Set orientation (Android Emulator only): portrait, landscape, or auto. Logged as an environment change.

Input parameters:

- `orientation` (string, required)
- `sessionId` (string, required)

### `qa_prepare_ios_target` (~193 tokens)

Prepare an iOS simulator target

Prepare an iOS Simulator: boot, install a simulator .app (consent-gated), launch, verify, and report whether WDA structured automation or visual-only mode is available. .ipa files are refused.

Input parameters:

- `allowLaunchWithoutMetro` (boolean): Launch a debug RN/Expo .app (no embedded main.jsbundle) without a serving Metro (it will RedBox).
- `app` (string): Simulator .app path (absolute or project-relative).
- `approve` (boolean)
- `attachWda` (string): auto (default): attach WDA if reachable, else visual-only; required: fail if it cannot attach; skip: visual-only.
- `bundleId` (string)
- `consentId` (string)
- `device` (string): Simulator UDID or exact name (ignoring case).
- `launch` (boolean)
- `sessionId` (string, required)

### `qa_prepare_target` (~187 tokens)

Prepare a target device + app

Prepare an Android Emulator and the app under test: bind or boot the device, Metro + adb reverse for RN/Expo, install, launch, verify. One combined consent covers boot/install; long steps return a jobId. bindOnly stops after bind/boot.

Input parameters:

- `allowLaunchWithoutMetro` (boolean): Launch a debug RN/Expo build without Metro (may RedBox).
- `apk` (string)
- `appId` (string)
- `approve` (boolean)
- `avd` (string)
- `bindOnly` (boolean): Bind/boot + adb reverse only.
- `consentId` (string)
- `device` (string): Target serial; required when >1 device is online.
- `force` (boolean)
- `headless` (boolean): Boot the AVD headless (default true).
- `sessionId` (string, required)

### `qa_report` (~139 tokens)

Build a session report

Build the session report: returns a short summary + URIs (reportMdUri = report.md to read; reportUri = full JSON, inline with format:"json" minus steps; reportHtmlPath = report.html, give the user that path). format adds an export: markdown, json, junit, sarif, github-summary, flow, playwright (deprecated). CI: docs/ci-reports.md.

Input parameters:

- `baseline` (string): Baseline report.json path; adds comparison links.
- `format` (string)
- `sessionId` (string, required)
- `trendRoot` (string): Project root with run history; adds trends/flakes.

### `qa_resolve_artifact` (~147 tokens)

Resolve a build artifact

Find the best installable build (.apk/.aab/.ipa/.app) in Gradle/Flutter/Xcode outputs. Returns ranked candidates (build type, installability, app id, warnings) and the paths searched. Side-effect free.

Input parameters:

- `allowOutsideRoot` (boolean): Allow a candidate outside the project root.
- `buildType` (string)
- `path` (string): Explicit artifact path; skips the search.
- `platform` (string)
- `projectRoot` (string): Absolute path; else the usual root resolution.
- `requireInstallableOn` (string)
- `sessionId` (string): Use this session's projectRoot if given.

### `qa_resolve_target` (~163 tokens)

Resolve the best test target

Pick the emulator/simulator to test on, deterministically (honors platform, device, artifact; prefers an online one, else plans a boot). Returns selected, reason, alternatives, willBoot. include: context (framework, artifacts, devices, toolchain) and/or plan (READY/BLOCKED/UNSAFE workflows). Boots nothing; physical phones are unsupported.

Input parameters:

- `device` (string): Explicit adb serial / simulator udid or name / AVD name.
- `include` (array): Extra sections: context, plan.
- `platform` (string)
- `preferRealDevice` (boolean): Refused (PHYSICAL_DEVICE_UNSUPPORTED) when a phone is visible.
- `projectRoot` (string)
- `sessionId` (string)

### `qa_screen_record` (~119 tokens)

Record the screen

Record the screen to an mp4 artifact: start (consent-gated; avoid password/OTP screens), status, stop. save:"on_failure" keeps only failing runs. One recording per session; Android stops after ~3 min.

Input parameters:

- `action` (string, required)
- `approve` (boolean)
- `consentId` (string)
- `failed` (boolean): stop with on_failure: true keeps the video.
- `save` (string): start: always (default) | on_failure.
- `sessionId` (string, required)

### `qa_screenshot` (~120 tokens)

Capture a screenshot

Save a screenshot as a session artifact and return its swipium:// URI. Withheld on password/OTP screens unless force:true. Counts against the screenshot budget. inline:true also returns the image (JPEG, long edge 1568 px, scale in the text) for clients that can see images.

Input parameters:

- `force` (boolean)
- `inline` (boolean): Also return the image inline (the result then has no structuredContent).
- `reason` (string): What it documents (shown in qa_report).
- `sessionId` (string, required)

### `qa_smoke` (~126 tokens)

Run a smoke test

Run a smoke test server-side: optionally launch, run the baseline (snapshot quality + health + evidence screenshot), then every saved flow in .swipium/flows. Records a qa_note per workflow; call qa_report after. Needs a prepared device.

Input parameters:

- `launch` (boolean): Launch first (default true with an appId).
- `runFlows` (boolean): Run saved .swipium/flows (default true).
- `sessionId` (string, required)
- `variables` (object): ${VAR} values for flows (over stored inputs, SWIPIUM_* env).

### `qa_snapshot` (~135 tokens)

Snapshot the screen

Read the current screen as addressable elements (@e1, @e2...) for qa_act, with a snapshotQuality verdict; call it before acting and after navigation. Interactive elements only, capped on busy screens (filter finds the rest). A trailing " -" marks a non-clickable element. Refs stay stable while the element stays.

Input parameters:

- `diff` (boolean): Same screen: only added/removed/changed refs since the previous snapshot; a new screen returns the full list.
- `filter` (string): Substring match on text/label/id/role (finds capped elements).
- `sessionId` (string, required)

Output parameters:

- `diff`
- `elementCount` (number)
- `elements` (array)
- `ok` (boolean)
- `screen` (string)

### `qa_start_session` (~202 tokens)

Start a QA session

Open a QA session for the low-level tools (qa_snapshot, qa_act, flows); qa_test_this opens its own. projectRoot defaults to MCP roots > SWIPIUM_PROJECT_ROOT > CLAUDE_PROJECT_DIR > a project-like cwd. Default budget: 30 min / 60 actions / 40 screenshots. fixtures declare preconditions; unmet ones report blocked, not failed.

Input parameters:

- `budget` (object): Override default budget caps.
- `fixtures` (array): Preconditions, merged with .swipium/fixtures.json: {name, requiredState?, ...} (docs/tools.md#qa_start_session).
- `profile` (string): Budget class.
- `projectRoot` (string): Absolute app path.
- `responseMode` (string): Output detail: compact (summary + URIs), normal (default, + JSON), verbose (all JSON, element objects).
- `sensitive` (boolean): Refuse all screenshots, recordings, and on-screen evidence.

### `qa_status` (~126 tokens)

Status, orientation, and next step

Session status and the next tool to call; cheap, call between steps. Without sessionId: first-call orientation (how to drive Swipium: first call, polling, stop rules, tool groups). With sessionId: compact state (device, app, budget left, counters, findings, last job, workarounds, readiness) and nextBestAction (tool, args, why); goal biases it.

Input parameters:

- `goal` (string): Bias nextBestAction toward this autopilot goal.
- `sessionId` (string): Omit for orientation; pass to get that session state.

### `qa_suite_export` (~67 tokens)

Export the persistent QA test suite

Export the repo test suite as markdown, yaml, json, or junit; save:true writes it under .swipium/test-suite-export/.

Input parameters:

- `format` (string, required)
- `projectRoot` (string)
- `save` (boolean)
- `sessionId` (string)

### `qa_suite_generate` (~135 tokens)

Generate persistent suite cases from the app

Generate cases from this session (recorded actions, outcomes, exploration coverage) and merge them into the durable repo suite .swipium/test-suite.json (re-runs update, no duplicates). Returns cases, skipped features, and coverage gaps. Per-run assets: qa_generate.

Input parameters:

- `creativity` (string): How far beyond the happy path (default standard).
- `creativityLevel` (string): Deprecated alias of creativity.
- `feature` (string): Functionality label (default: the recorded flow name).
- `includeManualOnly` (boolean)
- `projectRoot` (string)
- `sessionId` (string)

### `qa_suite_lint` (~93 tokens)

Lint the persistent QA test suite

Lint the repo test suite (missing results, stale links, duplicate ids, brittle automation, unsafe adversarial cases) and generated page objects (fragile locators). Returns errors and warnings.

Input parameters:

- `brittleThreshold` (string)
- `liveFeatureIds` (array): Live map feature ids (enables the stale-link rule).
- `projectRoot` (string)
- `sessionId` (string)

### `qa_suite_read` (~89 tokens)

Read the persistent QA test suite

Read the repo suite (.swipium/test-suite.json), filtered by functionality/status, as a summary, json, or markdown, plus a resource URI for the full suite.

Input parameters:

- `format` (string)
- `functionality` (string): Only return cases under this functionality/feature.
- `projectRoot` (string)
- `sessionId` (string)
- `status` (string)

### `qa_suite_update` (~100 tokens)

Update the persistent QA test suite

Add or update cases in the repo test suite: a case matching feature + objective + steps is updated, new ones get stable TC ids. mergeMode decides whether generated fields overwrite curated ones.

Input parameters:

- `cases` (array): Canonical or partial cases (normalized).
- `mergeMode` (string)
- `projectRoot` (string)
- `sessionId` (string)
- `source` (string, required)
- `sourceUri` (string)

### `qa_test_feature` (~308 tokens)

Test a feature (focused)

Test one named feature (e.g. login, checkout). mode:"execute" (default): a job that explores toward it, records pass/fail/blocked per case, and writes a report (poll qa_job_status). mode:"plan" (read-only): scope, cases, fixtures, plan. Without sessionId it bootstraps from projectRoot (consent-gated). Auth, paywall, or missing-fixture gates report blocked, not failed.

Input parameters:

- `allowAdversarial` (boolean)
- `approve` (boolean)
- `consentId` (string): Bootstrap consent (boot/install/launch).
- `creativity` (string)
- `device` (string): Device/simulator to bootstrap on.
- `feature` (string, required): The feature to test, in natural language.
- `generateCases` (boolean): Generate test cases (default true).
- `includeCode` (boolean)
- `limit` (number)
- `maxActions` (number): Default 20.
- `maxScreens` (number): Default 8.
- `mode` (string): execute (default, job) | plan (read-only) | interactive (until the first question).
- `platform` (string)
- `projectRoot` (string)
- `sessionId` (string): Prepared session (else bootstrap from projectRoot).
- `timeoutMs` (number): Ignored (kept for compatibility): the run is bounded by maxScreens / maxActions; poll qa_job_status.

### `qa_test_this` (~458 tokens)

Test this app (autopilot)

Start here to test a mobile app (Android/iOS): finds or builds the app, picks an emulator/simulator, runs prepare > smoke > explore > report > suite as a job (mode:"execute", default; pass waitForCompletion:true or poll qa_job_status). mode:"plan" only plans, no side effects. One consent covers boot/install/build; every terminal state has a report. iOS without WDA: visual-only smoke.

Input parameters:

- `allowOutsideRoot` (boolean)
- `approve` (boolean)
- `buildIfNeeded` (boolean): Build from source when no artifact exists (default true).
- `consentId` (string)
- `device` (string)
- `explore` (boolean): execute: run guided exploration after the smoke.
- `fastSmoke` (boolean): Launch + smoke only, no suite (ignored with goal/generateSuite).
- `generateSuite` (boolean): execute: also generate a POM suite from the run.
- `goal` (string): Default: smoke, then a suite ("smoke" is fastest). Presets explore/generateSuite/stopOnNeedsInput; explicit flags win.
- `goalText` (string): reproduce_bug: the bug/flow to focus on.
- `mode` (string): execute (default, job) | plan (no side effects) | interactive (until the first question).
- `platform` (string): Force a platform (default inferred).
- `preferRealDevice` (boolean): Out of scope: returns PHYSICAL_DEVICE_UNSUPPORTED.
- `projectRoot` (string)
- `responseMode` (string): compact = summary + URIs only; kept as the session default for later calls.
- `sessionId` (string): Reuse a session.
- `stopOnNeedsInput` (boolean): execute: ask for login/test data instead of testing pre-login only.
- `timeoutMs` (integer): waitForCompletion window (default 45000, max 50000; SWIPIUM_MAX_WAIT_MS raises both; larger values are clamped); still running after it > state:"running" + jobId, poll qa_job_status.
- `waitForCompletion` (boolean): execute: block until done (or timeoutMs).

Output parameters:

- `consentId` (string)
- `jobId` (string)
- `mode` (string)
- `ok` (boolean)
- `reportUri` (string)
- `requiresConsent` (boolean)
- `sessionId` (string)
- `state` (string): running | completed | blocked | unsafe | needs_input | ready

### `qa_visual` (~337 tokens)

Visual assert / baseline / diff / find text / find image

Screenshot checks where the UI tree fails (maps, canvases, WDA-less iOS). mode: assert, baseline (.swipium/baselines/<name>.png), diff, find_text (OCR, consent-gated), find_image. Finds return device coordinates + coordinateSpace; tap:true taps it (recorded + budgeted). Password/OTP screens need force:true; unverified credential screens: assert/diff not saved. Details: docs/tools.md#qa_visual.

Input parameters:

- `approve` (boolean)
- `assertion` (string): assert: what you visually confirmed.
- `consentId` (string)
- `force` (boolean): Capture even if a secure field may be visible.
- `minConfidence` (number): find_text: min OCR confidence 0..1 (default 0.8).
- `minScore` (number): find_image: min match score 0..1 (default 0.85).
- `mode` (string, required)
- `name` (string): baseline/diff: baseline name, [A-Za-z0-9._-]{1,64}.
- `pass` (boolean): assert: default true; false if the expected result is NOT visible.
- `query` (string): find_text: text to find.
- `reason` (string): assert: extra detail.
- `sessionId` (string, required)
- `tap` (boolean): find_text/find_image: tap the found point.
- `template` (string): find_image: project PNG or swipium:// URI.
- `threshold` (number): diff: max changed fraction to pass (default 0.02).

### `qa_wait` (~135 tokens)

Wait for a setup condition

Wait for a setup condition instead of shell sleep: device_online (adb), metro_ready, wda_ready (after qa_wda start returned status:"starting"), or simulator_booted (the bound iOS Simulator, after qa_ios boot returned status:"booting"). Returns satisfied, or timedOut:true (call again). Jobs: qa_job_status waitMs.

Input parameters:

- `for` (string, required)
- `sessionId` (string, required)
- `timeoutMs` (integer): default 45000; values above SWIPIUM_MAX_WAIT_MS (default 50000) are clamped. On timedOut, call again to keep waiting.

### `qa_wda` (~367 tokens)

iOS WebDriverAgent setup and control

Set up and manage iOS WebDriverAgent (WDA), needed for structured iOS tap/type/snapshot; start with action:"setup": it reuses a running WDA, else builds (if no build exists) and starts it as one consent-gated job (poll qa_job_status, then qa_wda attach). build/start are consent-gated; attach takes an external webDriverAgentUrl; status/doctor/diagnose/logs/tune inspect.

Input parameters:

- `action` (string, required): build: background job, reuses a build unless rebuild:true. start: waits up to 45 s, then status:"starting" (poll qa_wait for:"wda_ready").
- `allowNonLoopback` (boolean): Required to use a non-loopback external WDA URL.
- `approve` (boolean)
- `bundleId` (string): Default: the session appId.
- `consentId` (string)
- `derivedDataPath` (string): xcodebuild -derivedDataPath for reuse.
- `device` (string): Simulator UDID (default: session device).
- `rebuild` (boolean): build/setup: run build-for-testing even when a previous build (.xctestrun) exists.
- `scheme` (string): Default WebDriverAgentRunner.
- `sessionId` (string, required)
- `udid` (string): Deprecated alias of device.
- `wdaProjectPath` (string): WebDriverAgent.xcodeproj (default: Appium WDA if found).
- `webDriverAgentUrl` (string): Default: this session's WDA (attached, or the last start / reuse), else ios.wda.url (http://127.0.0.1:8100). Non-loopback needs allowNonLoopback + consent.

## Diagnostics

Captured diagnostic sections: Provenance, Dependencies. The full working is on the page: https://verifymcp.io/servers/geropalombo-swipium/swipium#diagnostics

## Score history

- 2026-11-04: 90
- 2026-11-03: 89
- 2026-11-02: 74
- 2026-11-01: 88
- 2026-10-31: 88
- 2026-10-30: 87
- 2026-10-29: 86
- 2026-10-28: 86
- 2026-10-27: 86
- 2026-10-26: 85
- 2026-10-25: 85
- 2026-10-24: 84
- 2026-10-23: 84
- 2026-10-22: 84
- 2026-10-21: 83
- 2026-10-20: 83
- 2026-10-19: 67
- 2026-10-18: 70
- 2026-10-17: 70
- 2026-10-16: 70
- 2026-10-15: 70
- 2026-10-14: 70
- 2026-10-13: 40
- 2026-10-12: 40

## Common questions

### What is the Swipium MCP server?

Swipium is an MCP server listed in the public MCP registry as io.github.GeroPalombo/swipium. Mobile QA on local Android Emulators and iOS Simulators: autopilot tests, evidence, reports. This page covers its npm package (swipium).

### Is the Swipium MCP server safe to use?

Swipium scores 90 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 4 November 2026. It declares no install or post-install scripts. Its build provenance is signed and verified. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the Swipium MCP server expose?

Swipium exposes 55 tools: qa_act, qa_app_control, qa_app_map_build, qa_app_map_feature_scope, qa_app_map_query, and 50 more. Their descriptions and schemas cost roughly 9,344 tokens of context every time the server is loaded.

### Is the Swipium MCP server still maintained?

Swipium is still listed as active in the MCP registry. We last reached this channel on 4 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

### What licence is the Swipium MCP server under?

Swipium declares the MIT licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.

## Links

- npm package: https://www.npmjs.com/package/swipium
- Socket report: https://socket.dev/npm/package/swipium
- Repository: https://github.com/GeroPalombo/swipium
- Website: https://swipium.com/
- Changelog RSS feed: https://verifymcp.io/servers/geropalombo-swipium/swipium.xml
- Changelog JSON feed: https://verifymcp.io/servers/geropalombo-swipium/swipium.json
- HTML version of this page: https://verifymcp.io/servers/geropalombo-swipium/swipium