yaps-audio-cleaner

• Outcome: Review
• Latest audit: Sep 25, 2026 · 1:28 AM UTC
• Version: 1.0.2

Overview

The skill is advertised as an audio cleaner, but its bundled runtime exposes broader Yaps CLI, transcription, meeting, auth, and app-launch behavior that users should review before installing.

Install only if you are comfortable with a local skill invoking the installed Yaps CLI beyond a single denoise command, checking sanitized Yaps account status, and possibly opening Yaps to refresh account state. Treat it as a broader Yaps automation adapter packaged as an audio-cleaning workflow, and review any requested command, input file, output path, model download, or overwrite action before approving it.

A.I.G

Vulnerability Patterns

• Skill Instruction Hijacking: Alters the agent's session goals or safety constraints when the skill loads
• Agent Memory Poisoning: Writes attacker-controlled rules into memory that affect later sessions
• Remote Payload Retrieval and Execution: Fetches external code whose behavior can change after review
• Embedded Malicious Code: Ships malicious scripts inside the skill and executes them locally
• Unauthorized Access and Privilege Escalation: Obtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Rogue Agent: Self-Modification, Session Persistence
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection

Findings (22)

Context-Inappropriate Capability

High

• Category: Not specified by scanner
• Confidence: 97% confidence

Finding

The adapter supports broad passthrough of arbitrary Yaps CLI arguments and request-mode arrays, allowing callers to invoke functionality far beyond the stated audio-cleaning use case. In the context of an agent skill, this creates a confused-deputy risk where the skill becomes a generic wrapper around the local CLI, potentially exposing unexpected data-processing, file-writing, or account-backed operations not reviewed by the user.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High

• Category: Not specified by scanner
• Confidence: 99% confidence

Finding

The skill metadata promises audio cleaning, but the implementation performs transcription, subtitle generation, meeting transcription, and media extraction. This capability mismatch is dangerous because a host or user may grant file access under the assumption of simple denoising while the code can process and emit speech content and derived artifacts, expanding access to sensitive audio and text data beyond the declared purpose.

Content

No source excerpt is available for this finding.

Credential Access

High

• Category: Privilege Escalation
• Confidence: 70% confidence

Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · runtime/yaps-cli-discovery.mjs (reported line 538)May include surrounding context.

js

function accountStatusUnsafeDiagnosis(appVersion) {
  return {
    code: "account_status_unsafe",
    message: `Yaps ${appVersion || "before 2.3.124"} uses an older credential-based account check, so this plugin deliberately did not run it. Update Yaps to 2.3.124 or newer; the plugin will then reuse the desktop sign-in, trial, or Yaps Pro automatically. Do not approve a Keychain prompt or create a separate plugin account.`,
  };
}

Credential Access

High

• Category: Privilege Escalation
• Confidence: 70% confidence

Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · runtime/yaps-cli-discovery.mjs (reported line 1024)May include surrounding context.

js

function accountStatusUnsafeDiagnosis(appVersion) {
  return {
    code: "account_status_unsafe",
    message: `Yaps ${appVersion || "before 2.3.124"} uses an older credential-based account check, so this plugin deliberately did not run it. Update Yaps to 2.3.124 or newer; the plugin will then reuse the desktop sign-in, trial, or Yaps Pro automatically. Do not approve a Keychain prompt or create a separate plugin account.`,
  };
}

Credential Access

High

• Category: Privilege Escalation
• Confidence: 70% confidence

Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · runtime/yaps-cli-discovery.mjs (reported line 1035)May include surrounding context.

js

function accountStatusUnsafeDiagnosis(appVersion) {
  return {
    code: "account_status_unsafe",
    message: `Yaps ${appVersion || "before 2.3.124"} uses an older credential-based account check, so this plugin deliberately did not run it. Update Yaps to 2.3.124 or newer; the plugin will then reuse the desktop sign-in, trial, or Yaps Pro automatically. Do not approve a Keychain prompt or create a separate plugin account.`,
  };
}

Credential Access

High

• Category: Privilege Escalation
• Confidence: 70% confidence

Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · runtime/yaps-cli-discovery.mjs (reported line 1064)May include surrounding context.

js

function accountStatusUnsafeDiagnosis(appVersion) {
  return {
    code: "account_status_unsafe",
    message: `Yaps ${appVersion || "before 2.3.124"} uses an older credential-based account check, so this plugin deliberately did not run it. Update Yaps to 2.3.124 or newer; the plugin will then reuse the desktop sign-in, trial, or Yaps Pro automatically. Do not approve a Keychain prompt or create a separate plugin account.`,
  };
}

Context-Inappropriate Capability

High

• Category: Not specified by scanner
• Confidence: 98% confidence

Finding

The module can launch the installed Yaps desktop application in the background to refresh account state. Spawning local GUI applications is a powerful side effect unrelated to denoising and can surprise users, interact with their session, and be abused to trigger unintended actions or data access in the trusted desktop app context.

Content

No source excerpt is available for this finding.

Credential Access

High

• Category: Privilege Escalation
• Confidence: 70% confidence

Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · runtime/yaps-cli-discovery.mjs (reported line 1061)May include surrounding context.

js

message: "Yaps is signed in, but its trial or Yaps Pro access is not active. Open Yaps to review the available trial or Yaps Pro options; the plugin will pick up the change automatically.",
    };
  }
  if (auth.status === "credential_unavailable" || auth.diagnosticCode === "keychain_unavailable") {
    return {
      code: "account_status_unsupported",
      message: "This installed Yaps helper uses an older credential-based account check. Update Yaps and retry. Do not approve a Keychain prompt or create a separate plugin account.",

Undeclared Tool Scope

Medium

• Category: MCP Least Privilege
• Confidence: 70% confidence

Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium

• Category: Not specified by scanner
• Confidence: 72% confidence

Finding

The comment states the adapter writes no settings, yet prepareInvocation resolves a Yaps session and passes session.settingsPath into applyResolvedSettings for command execution. Even if the adapter itself may not directly persist settings, the documentation's blanket claim is stronger than the actual behavior and can mislead about settings-related side effects.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium

• Category: Not specified by scanner
• Confidence: 94% confidence

Finding

The meeting workflow creates transcription projects and returns internal project_path and audio_path values, which are unrelated to audio cleaning and may reveal local filesystem layout and locations of persisted media. While not direct code execution, this unnecessarily broadens data exposure and can help an agent or downstream consumer discover and access sensitive local artifacts.

Content

No source excerpt is available for this finding.

Session Persistence

Medium

• Category: Rogue Agent
• Confidence: 75% confidence

Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · runtime/yaps-cli-discovery.mjs (reported line 13)May include surrounding context.

js

export const MIN_SAFE_AUTH_STATUS_VERSION = "2.3.124";
const MAX_PROBE_BYTES = 64 * 1024;
const MAX_VERSION_BYTES = 4 * 1024;
const MAX_PLIST_BYTES = 64 * 1024;
const MAX_WINDOWS_SHIM_BYTES = 4 * 1024;
const RUNNING_YAPS_PROCESS_NAMES = new Set(["yaps.exe", "yaps_mcp.exe", "yaps_cli.exe"]);
// Fixed WMI filter: never interpolate a discovered path into this command.

Session Persistence

Medium

• Category: Rogue Agent
• Confidence: 75% confidence

Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · runtime/yaps-cli-discovery.mjs (reported line 422)May include surrounding context.

js

export const MIN_SAFE_AUTH_STATUS_VERSION = "2.3.124";
const MAX_PROBE_BYTES = 64 * 1024;
const MAX_VERSION_BYTES = 4 * 1024;
const MAX_PLIST_BYTES = 64 * 1024;
const MAX_WINDOWS_SHIM_BYTES = 4 * 1024;
const RUNNING_YAPS_PROCESS_NAMES = new Set(["yaps.exe", "yaps_mcp.exe", "yaps_cli.exe"]);
// Fixed WMI filter: never interpolate a discovered path into this command.

Context-Inappropriate Capability

Medium

• Category: Not specified by scanner
• Confidence: 93% confidence

Finding

The code enumerates running Windows processes via PowerShell/WMI to discover Yaps executable paths. Even though it uses a fixed filter and avoids command injection, process enumeration is a host-inspection capability that exceeds what is necessary for simple audio cleaning and can expose local environment details.

Content

No source excerpt is available for this finding.

Session Persistence

Medium

• Category: Rogue Agent
• Confidence: 75% confidence

Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · runtime/yaps-cli-discovery.mjs (reported line 419)May include surrounding context.

js

if (platform === "darwin") {
    const application = macApplicationForCli(cli.path, canonicalize);
    if (!application) return null;
    const plist = posix.join(application, "Contents", "Info.plist");
    try {
      const metadata = statFile(plist);
      if (!metadata.isFile() || metadata.size > MAX_PLIST_BYTES) return null;

Session Persistence

Medium

• Category: Rogue Agent
• Confidence: 75% confidence

Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · runtime/yaps-cli-discovery.mjs (reported line 421)May include surrounding context.

js

if (platform === "darwin") {
    const application = macApplicationForCli(cli.path, canonicalize);
    if (!application) return null;
    const plist = posix.join(application, "Contents", "Info.plist");
    try {
      const metadata = statFile(plist);
      if (!metadata.isFile() || metadata.size > MAX_PLIST_BYTES) return null;

Session Persistence

Medium

• Category: Rogue Agent
• Confidence: 75% confidence

Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · runtime/yaps-cli-discovery.mjs (reported line 427)May include surrounding context.

js

if (platform === "darwin") {
    const application = macApplicationForCli(cli.path, canonicalize);
    if (!application) return null;
    const plist = posix.join(application, "Contents", "Info.plist");
    try {
      const metadata = statFile(plist);
      if (!metadata.isFile() || metadata.size > MAX_PLIST_BYTES) return null;

Session Persistence

Medium

• Category: Rogue Agent
• Confidence: 75% confidence

Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · runtime/yaps-cli-discovery.mjs (reported line 434)May include surrounding context.

js

if (platform === "darwin") {
    const application = macApplicationForCli(cli.path, canonicalize);
    if (!application) return null;
    const plist = posix.join(application, "Contents", "Info.plist");
    try {
      const metadata = statFile(plist);
      if (!metadata.isFile() || metadata.size > MAX_PLIST_BYTES) return null;

Session Persistence

Medium

• Category: Rogue Agent
• Confidence: 75% confidence

Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · runtime/yaps-cli-discovery.mjs (reported line 438)May include surrounding context.

js

if (platform === "darwin") {
    const application = macApplicationForCli(cli.path, canonicalize);
    if (!application) return null;
    const plist = posix.join(application, "Contents", "Info.plist");
    try {
      const metadata = statFile(plist);
      if (!metadata.isFile() || metadata.size > MAX_PLIST_BYTES) return null;

Session Persistence

Medium

• Category: Rogue Agent
• Confidence: 75% confidence

Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · runtime/yaps-cli-discovery.mjs (reported line 442)May include surrounding context.

js

if (platform === "darwin") {
    const application = macApplicationForCli(cli.path, canonicalize);
    if (!application) return null;
    const plist = posix.join(application, "Contents", "Info.plist");
    try {
      const metadata = statFile(plist);
      if (!metadata.isFile() || metadata.size > MAX_PLIST_BYTES) return null;

Context-Inappropriate Capability

Medium

• Category: Not specified by scanner
• Confidence: 95% confidence

Finding

The code reads and interprets Yaps authentication/account status, including alternate settings-path recommendations and desktop-account following behavior. For an audio-cleaning skill, inspecting auth state is a scope overreach that exposes sensitive local state and ties the skill to account/session introspection not required for processing audio.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium

• Category: Not specified by scanner
• Confidence: 96% confidence

Finding

This module goes beyond CLI discovery and performs session resolution, account-state handling, and recovery behavior. In an audio-cleaning skill, that scope expansion is risky because it touches local application state and authentication context unrelated to the user-visible denoising task, increasing privacy and attack surface.

Content

No source excerpt is available for this finding.