Back to skill

Security audit

Sonos 音乐点播

Security checks for vulnerabilities and agentic risk

Overview

This Sonos playback skill is coherent and disclosed, but users should notice that it installs an unpinned Python dependency and relies on local workspace playback scripts not included in the package.

Install only if you are comfortable letting the skill create a local Python venv, install soco from pip, access Sonos devices on your local network, and use the referenced workspace playback scripts. Review or supply the missing playback scripts yourself, and prefer pinned dependencies if you need stronger reproducibility.

Vulnerability Patterns
  • 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
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (1)

T08 · Insecure Dependencies

Warning
Location
scripts/bootstrap_env.sh:38
Finding

Unpinned Third-Party Package Installation from an Implicit Package Index

Content
View full analysis
/dev/null 2>&1; then "$PIP_BIN" install --upgrade pip soco fi ``` ### Technical Analysis The bootstrap process installs `pip` and `soco` without fixed versions, integrity hashes, a lock file, or an explicitly trusted package index. Consequently, the code installed by an otherwise unchanged Skill can vary over time or according to the user's `pip` configuration. The command also upgrades `pip`, expanding the mutable dependency set executed during installation. Python packages and their build dependencies may execute code during package building or installation. The installed `soco` package is subsequently imported by the bootstrap process, which executes its module initialization code. Exploitation requires control or compromise of a relevant package release, transitive dependency, configured package index, package mirror, or dependency-resolution path. The reviewed code does not itself establish such control, but it lacks controls that would prevent a supply-chain compromise from reaching the local environment. ### Attack Path 1. An attacker compromises a package release, transitive dependency, package mirror, or package index used by `pip`. 2. The user runs `scripts/install.sh`, which invokes `scripts/bootstrap_env.sh`. 3. If `soco` cannot already be imported from the virtual environment, the bootstrap executes: ```sh "$PIP_BIN" install --upgrade pip soco ``` 4. `pip` resolves and installs the attacker-controlled or compromised package without checking an audited version or expected hash. 5. Malicious code may execute during package build or installation. 6. The bootstrap then imports `soco` when printing its version, and later playback components may import i ...[truncated 769 chars]
Remediation
View remediation
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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 (2)

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
92% confidence
Finding

The skill documents and expects shell execution (install.sh, check_env.sh, bootstrap_env.sh, wrapper scripts, and Sonos CLI usage) but does not declare an explicit tool scope such as permissions or allowed-tools. This creates an authorization gap where an agent may invoke shell actions without a clearly bounded contract, increasing the risk of unintended command execution or overbroad access in environments that rely on manifest-declared tool restrictions.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The description and overview explicitly state that the skill is intended for Chinese users ("面向中文用户") without indicating that language choice is optional or configurable. This creates a natural-language locale policy issue because it constrains usage to a specific language group without user opt-in or a clearly justified regional-compliance requirement.

Content

No source excerpt is available for this finding.