Back to skill

Security audit

Computer-Use Skill

Security checks for vulnerabilities and agentic risk

Overview

This is a real desktop-control skill, but it can grant broad computer-use access without a real user approval step and uses unfiltered screenshots.

Install only if you are comfortable giving an agent local desktop-control power. Keep sensitive windows and clipboard contents away from the session, and treat this build as needing review until request_access requires real user approval and Python dependencies are pinned and hash-verified.

Vulnerability Patterns
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Findings (2)

T05 · Unauthorized Access and Privilege Escalation

Error
Location
project/platforms/linux/src/session.ts:23
Finding

Application Access Requests Are Silently Auto-Approved

Content
View full analysis
app.resolved && !app.alreadyGranted) .map(app => ({ bundleId: app.resolved!.bundleId, displayName: app.resolved!.displayName, grantedAt: Date.now(), tier: app.proposedTier, })) const denied = req.apps .filter(app => !app.resolved) .map(app => ({ bundleId: app.requestedName, reason: 'not_installed' as const, })) return { granted, denied, flags: { ...DEFAULT_GRANT_FLAGS, ...req.requestedFlags, }, } } export function createSessionContext(): ComputerUseSessionContext { const state: State = { allowedApps: [], grantFlags: { ...DEFAULT_GRANT_FLAGS }, hiddenDuringTurn: new Set(), } return { getAllowedApps: () => state.allowedApps, getGrantFlags: () => state.grantFlags, getUserDeniedBundleIds: () => [], getSelectedDisplayId: () => state.selectedDisplayId, getDisplayPinnedByModel: () => state.displayPinnedByModel ?? false, getDisplayResolvedForApps: () => state.displayResolvedForApps, getLastScreenshotDims: () => state.lastScreenshotDims, onPermissionRequest: async req => autoApprovePermission(req), ``` ### Technical Analysis The MCP computer-use framework treats `onPermissionRequest` as the authorization boundary through which a host should display a consent request and wait for the user to approve or deny access. The standalone session implementation instead routes the request directly to `autoApprovePermis ...[truncated 2879 chars]
Remediation
View remediation

T08 · Insecure Dependencies

Warning
Location
project/platforms/linux/src/computer-use/pythonBridge.ts:73
Finding

Unpinned Python Dependencies Are Automatically Installed and Executed on First Use

Content
View full analysis
=10.1.0 Pillow>=11.3.0 pyautogui>=0.9.54 psutil>=7.0.0 python-xlib>=0.33 ``` The first-use bootstrap automatically upgrades pip and installs those mutable dependencies: ```ts const requirements = await readFile(requirementsPath, 'utf8') const digest = createHash('sha256').update(requirements).digest('hex') let installedDigest = '' try { installedDigest = (await readFile(installStampPath, 'utf8')).trim() } catch {} if (installedDigest !== digest) { logDebug('installing python runtime dependencies') await runOrThrow(pythonBinPath(), ['-m', 'pip', 'install', '--upgrade', 'pip'], 'pip upgrade') await runOrThrow( pythonBinPath(), ['-m', 'pip', 'install', '-r', requirementsPath], 'python dependency install', ) await writeFile(installStampPath, `${digest}\n`, 'utf8') } ``` ### Technical Analysis The requirements use lower-bound constraints rather than exact, reviewed versions. Dependency resolution can therefore select releases published after this Skill was audited. The installation also lacks package hashes, so it does not cryptographically constrain the downloaded artifacts to known files. The bootstrap upgrades pip before dependency installation. This adds another mutable component to the execution chain and causes package-management code that was not part of the reviewed Skill package to be downloaded and executed automatically. The SHA-256 stamp only hashes the local text of `requirements.txt`. It does not identify the versions that ...[truncated 2322 chars]
Remediation
View remediation
Vulnerability Patterns
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (175)

Tp4

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

If only macOS-specific logging declarations are represented while the skill claims broader standalone runtimes and platform-selection behavior, the documentation is overstating implemented scope. This primarily creates trust and review issues, though it is less directly exploitable than undeclared desktop-control functions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

If only macOS-specific logging declarations are represented while the skill claims broader standalone runtimes and platform-selection behavior, the documentation is overstating implemented scope. This primarily creates trust and review issues, though it is less directly exploitable than undeclared desktop-control functions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

If only macOS-specific logging declarations are represented while the skill claims broader standalone runtimes and platform-selection behavior, the documentation is overstating implemented scope. This primarily creates trust and review issues, though it is less directly exploitable than undeclared desktop-control functions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

If only macOS-specific logging declarations are represented while the skill claims broader standalone runtimes and platform-selection behavior, the documentation is overstating implemented scope. This primarily creates trust and review issues, though it is less directly exploitable than undeclared desktop-control functions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

If only macOS-specific logging declarations are represented while the skill claims broader standalone runtimes and platform-selection behavior, the documentation is overstating implemented scope. This primarily creates trust and review issues, though it is less directly exploitable than undeclared desktop-control functions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

If only macOS-specific logging declarations are represented while the skill claims broader standalone runtimes and platform-selection behavior, the documentation is overstating implemented scope. This primarily creates trust and review issues, though it is less directly exploitable than undeclared desktop-control functions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

If only macOS-specific logging declarations are represented while the skill claims broader standalone runtimes and platform-selection behavior, the documentation is overstating implemented scope. This primarily creates trust and review issues, though it is less directly exploitable than undeclared desktop-control functions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

If only macOS-specific logging declarations are represented while the skill claims broader standalone runtimes and platform-selection behavior, the documentation is overstating implemented scope. This primarily creates trust and review issues, though it is less directly exploitable than undeclared desktop-control functions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

If only macOS-specific logging declarations are represented while the skill claims broader standalone runtimes and platform-selection behavior, the documentation is overstating implemented scope. This primarily creates trust and review issues, though it is less directly exploitable than undeclared desktop-control functions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

If only macOS-specific logging declarations are represented while the skill claims broader standalone runtimes and platform-selection behavior, the documentation is overstating implemented scope. This primarily creates trust and review issues, though it is less directly exploitable than undeclared desktop-control functions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

If only macOS-specific logging declarations are represented while the skill claims broader standalone runtimes and platform-selection behavior, the documentation is overstating implemented scope. This primarily creates trust and review issues, though it is less directly exploitable than undeclared desktop-control functions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

If only macOS-specific logging declarations are represented while the skill claims broader standalone runtimes and platform-selection behavior, the documentation is overstating implemented scope. This primarily creates trust and review issues, though it is less directly exploitable than undeclared desktop-control functions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

If only macOS-specific logging declarations are represented while the skill claims broader standalone runtimes and platform-selection behavior, the documentation is overstating implemented scope. This primarily creates trust and review issues, though it is less directly exploitable than undeclared desktop-control functions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

If only macOS-specific logging declarations are represented while the skill claims broader standalone runtimes and platform-selection behavior, the documentation is overstating implemented scope. This primarily creates trust and review issues, though it is less directly exploitable than undeclared desktop-control functions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

If only macOS-specific logging declarations are represented while the skill claims broader standalone runtimes and platform-selection behavior, the documentation is overstating implemented scope. This primarily creates trust and review issues, though it is less directly exploitable than undeclared desktop-control functions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

If only macOS-specific logging declarations are represented while the skill claims broader standalone runtimes and platform-selection behavior, the documentation is overstating implemented scope. This primarily creates trust and review issues, though it is less directly exploitable than undeclared desktop-control functions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

If only macOS-specific logging declarations are represented while the skill claims broader standalone runtimes and platform-selection behavior, the documentation is overstating implemented scope. This primarily creates trust and review issues, though it is less directly exploitable than undeclared desktop-control functions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

If only macOS-specific logging declarations are represented while the skill claims broader standalone runtimes and platform-selection behavior, the documentation is overstating implemented scope. This primarily creates trust and review issues, though it is less directly exploitable than undeclared desktop-control functions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

If only macOS-specific logging declarations are represented while the skill claims broader standalone runtimes and platform-selection behavior, the documentation is overstating implemented scope. This primarily creates trust and review issues, though it is less directly exploitable than undeclared desktop-control functions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
88% confidence
Finding

If only macOS-specific logging declarations are represented while the skill claims broader standalone runtimes and platform-selection behavior, the documentation is overstating implemented scope. This primarily creates trust and review issues, though it is less directly exploitable than undeclared desktop-control functions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

If only macOS-specific logging declarations are represented while the skill claims broader standalone runtimes and platform-selection behavior, the documentation is overstating implemented scope. This primarily creates trust and review issues, though it is less directly exploitable than undeclared desktop-control functions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

If only macOS-specific logging declarations are represented while the skill claims broader standalone runtimes and platform-selection behavior, the documentation is overstating implemented scope. This primarily creates trust and review issues, though it is less directly exploitable than undeclared desktop-control functions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

If only macOS-specific logging declarations are represented while the skill claims broader standalone runtimes and platform-selection behavior, the documentation is overstating implemented scope. This primarily creates trust and review issues, though it is less directly exploitable than undeclared desktop-control functions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

If only macOS-specific logging declarations are represented while the skill claims broader standalone runtimes and platform-selection behavior, the documentation is overstating implemented scope. This primarily creates trust and review issues, though it is less directly exploitable than undeclared desktop-control functions.

Content

No source excerpt is available for this finding.

Missing User Warnings

High
Category
Not specified by scanner
Confidence
96% confidence
Finding

The README states that screenshot filtering is set to 'none', meaning captured screen content is not compositor- or runtime-filtered before being exposed to the MCP layer. For a desktop automation skill, this materially increases the risk of collecting highly sensitive on-screen information from unrelated applications, making the lack of a strong safety warning and operating guidance a real security concern.

Content

No source excerpt is available for this finding.