Piano Score Fingering Arrangement

• Outcome: Review
• Latest audit: Aug 16, 2026 · 8:30 AM UTC
• Version: 0.1.0

Overview

This skill is a local piano-score fingering tool that reads user-supplied score files and writes annotated outputs without hidden network, persistence, or unrelated actions.

Install only if you are comfortable giving the skill local access to the score files you provide and allowing it to create output PDFs, MusicXML, JSON plans, reports, and temporary files. Use trusted input PDFs where possible because the workflow preserves and overlays source PDF pages.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Behavioral AST: exec() Call, eval() Call, Dynamic Import
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration

Findings (6)

eval() call detected

High

• Category: Dangerous Code Execution

Content

raise NameError(arg)
        else:
            code = forward_ref.__forward_code__
            value = eval(code, globals, locals)
        forward_ref.__forward_evaluated__ = True
        forward_ref.__forward_value__ = value
        return value

• Confidence: 89% confidence
• Finding: This code evaluates forward-reference expressions with Python eval(), which can execute arbitrary code if annotations or forward reference strings are attacker-controlled. In many applications annotations are trusted developer input, but in systems that ingest untrusted Python objects, plugins, or generated code, this becomes a code-execution sink.

eval() call detected

High

• Category: Dangerous Code Execution

Content

locals = {param.__name__: param for param in type_params} | locals

        return_value = {key:
            value if not isinstance(value, str) else eval(value, globals, locals)
            for key, value in ann.items() }
        return return_value

• Confidence: 90% confidence
• Finding: get_annotations() evals string annotations when eval_str=True, which can execute arbitrary Python embedded in annotations. This is expected behavior for annotation resolution, but it is unsafe if used on untrusted modules, classes, or callables because annotation strings become a code-execution primitive.

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 92% confidence
• Finding: The skill instructs the host to read local files, write outputs, invoke Python scripts, and run shell commands, yet no explicit permissions are declared. This creates an authorization gap where a host or reviewer may underestimate the skill's real capabilities, increasing the chance of unintended file access or command execution beyond what users expect.

Context-Inappropriate Capability

Medium

• Confidence: 95% confidence
• Finding: The writer exposes an add_js() API that embeds document-opening JavaScript into generated PDFs. In a skill focused on piano score recognition and fingering, this active-content capability is unrelated to core functionality and could be abused to produce PDFs that trigger viewer actions, phishing dialogs, or other unsafe behavior when opened.

Context-Inappropriate Capability

Medium

• Confidence: 91% confidence
• Finding: The add_attachment() API allows arbitrary files to be embedded inside exported PDFs. That capability is not necessary for piano fingering annotation and can be misused to smuggle executables, scripts, or deceptive payloads inside otherwise trusted-looking score PDFs.

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: The code permits embedding open-on-launch JavaScript with no confirmation, warning, or policy enforcement at the API boundary. In this skill context, that increases the chance that generated sheet-music PDFs could carry hidden active behavior that users would not reasonably expect from a notation/fingering tool.

VirusTotal

63/63 vendors flagged this skill as clean.

Static analysis

Detected: suspicious.dynamic_code_execution, suspicious.exposed_secret_literal, suspicious.obfuscated_code

Dynamic code execution detected.

Critical

• Code: suspicious.dynamic_code_execution
• Location: scripts/vendor/typing_extensions.py:1485

File appears to expose a hardcoded API secret or token.

Critical

• Code: suspicious.exposed_secret_literal
• Location: scripts/vendor/pypdf/_encryption.py:629

File appears to expose a hardcoded API secret or token.

Critical

• Code: suspicious.exposed_secret_literal
• Location: scripts/vendor/pypdf/_reader.py:174

Potential obfuscated payload detected.

Warn

• Code: suspicious.obfuscated_code
• Location: scripts/vendor/pypdf/_doc_common.py:93

Potential obfuscated payload detected.

Warn

• Code: suspicious.obfuscated_code
• Location: scripts/vendor/pypdf/_encryption.py:112

Potential obfuscated payload detected.

Warn

• Code: suspicious.obfuscated_code
• Location: scripts/vendor/pypdf/generic/_image_inline.py:194