Global Video Title Generator

• Outcome: Pass
• Latest audit: May 28, 2026 · 2:29 PM UTC
• Version: 1.0.0

Overview

This is an over-broad but visible video title/content generator, with no evidence of hidden execution, exfiltration, destructive behavior, or active payment/network code.

Install only if you are comfortable with a creator content tool that goes beyond title generation. Do not enter real secrets or sensitive unpublished plans into any paid/API workflow until the provider and privacy terms are verified, and do not run setup.py or publish.py unless you want it to create or overwrite local project files.

SkillSpector

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands

Findings (17)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 91% confidence
• Finding: The skill advertises itself as a title generator, but the analyzed behavior indicates file read/write capabilities without any declared permissions. Undeclared filesystem access expands the trust boundary and can enable unintended access to local data or modification of workspace files, especially when users believe the skill is only performing remote title generation.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 96% confidence
• Finding: This is a substantial description-behavior mismatch: the skill claims to generate video titles, but the detected behavior includes broader content generation, file creation/publication tooling, monetization logic, and misleading language-support claims. Such mismatches are dangerous because they conceal materially different capabilities from users and reviewers, making it easier for a skill to perform unexpected actions or exfiltrate/modify data under a benign-looking label.

Description-Behavior Mismatch

Medium

• Confidence: 96% confidence
• Finding: The manifest identifies the skill as a video title generator in the provided context, but the file declares and documents a broader video content generator. This mismatch can cause callers, reviewers, or policy gates to trust the skill for a narrower capability while it actually solicits or produces additional outputs, increasing the risk of over-privileged use, incorrect routing, and unsafe downstream assumptions.

Description-Behavior Mismatch

Medium

• Confidence: 95% confidence
• Finding: The documentation advertises complete content generation features that exceed the stated skill purpose of title generation. In an agent ecosystem, this is dangerous because orchestration, approval, and user consent may be based on the narrower declared scope, allowing the skill to be invoked in situations where broader content generation was not expected or reviewed.

Description-Behavior Mismatch

Medium

• Confidence: 94% confidence
• Finding: The use cases and examples explicitly broaden the skill from title generation into descriptions, chapters, tags, and other content artifacts. This creates capability confusion for users and automated systems, which may lead to unauthorized processing, unintended content generation, or bypass of governance that applies differently to richer generated outputs.

Description-Behavior Mismatch

Medium

• Confidence: 92% confidence
• Finding: The config declares a broader 'global-video-content-generator' product with descriptions, tags, engagement prompts, chapters, pricing, payments, and growth strategy, which materially exceeds the advertised skill scope of a title-only generator. This scope mismatch is dangerous because it can conceal undeclared capabilities and business logic from reviewers and users, undermining trust boundaries and increasing the chance that downstream components expose features, collect data, or trigger monetization flows that were not authorized for this skill.

Description-Behavior Mismatch

Medium

• Confidence: 91% confidence
• Finding: The skill metadata says it is a video title generator, but the configuration defines a broader 'global-video-content-generator' that also produces descriptions, tags, engagement prompts, and chapters. This scope mismatch is dangerous because users and platform controls may grant trust or permissions based on a narrow purpose while the skill is actually designed for materially broader behavior.

Description-Behavior Mismatch

Medium

• Confidence: 95% confidence
• Finding: The monetization section introduces subscription tiers, payment providers, credit systems, and refund handling that go well beyond a simple title-generation skill description. Hidden commercial and payment-related capabilities increase risk because they can enable undisclosed billing flows, account abuse, or collection of financial and usage data outside the user’s expected interaction scope.

Description-Behavior Mismatch

Medium

• Confidence: 93% confidence
• Finding: The analytics block enables generation-quality tracking and user-behavior tracking, which are not disclosed by the title-generator description. This is risky because it expands the skill from content generation into telemetry collection, creating privacy and transparency concerns and increasing the chance of undisclosed data processing.

Context-Inappropriate Capability

Medium

• Confidence: 94% confidence
• Finding: User behavior tracking is not necessary for generating video titles and appears unrelated to the stated purpose of the skill. In this context, the mismatch makes the tracking more dangerous because it suggests unnecessary collection of behavioral data under the cover of a simple content tool.

Description-Behavior Mismatch

High

• Confidence: 97% confidence
• Finding: The file clearly implements more than title generation: it generates complete video content, descriptions, batch outputs, and usage/credit tracking. This mismatch between the manifest/skill description and actual behavior is dangerous because it expands the skill's capability beyond what reviewers and users would reasonably expect, undermining least-privilege review and informed consent.

Intent-Code Divergence

Medium

• Confidence: 94% confidence
• Finding: The module docstring explicitly states that it combines title generation with full content creation, which contradicts the declared skill purpose as a video title generator. In the context of an agent skill, this discrepancy is security-relevant because hidden or under-disclosed functionality can bypass appropriate scrutiny and lead to unexpected data processing or broader operational impact.

Description-Behavior Mismatch

Medium

• Confidence: 83% confidence
• Finding: This setup script performs broad project scaffolding, creates marketing and business-plan artifacts, and writes user-visible files that are not necessary for a video-title-generation skill. That scope mismatch increases supply-chain and trust risk because the package does more than the manifest implies, making users more likely to run unexpected code with filesystem write access.

Description-Behavior Mismatch

Medium

• Confidence: 79% confidence
• Finding: The script creates a user config template containing an API key field despite the manifest describing a title-generation skill with no clear need for secret handling in setup. Introducing credential-related configuration outside the declared scope can normalize secret collection and expand the attack surface for accidental disclosure or later misuse.

Context-Inappropriate Capability

Medium

• Confidence: 84% confidence
• Finding: The setup routine exposes revenue targets and business-model planning data unrelated to the advertised functionality. While not directly exploitable code execution, unjustified business-planning behavior in a setup script is suspicious and increases risk that users execute unnecessary logic under the guise of installation.

Missing User Warnings

Low

• Confidence: 82% confidence
• Finding: The examples encourage use of an API key and remote API endpoints without clearly warning that user inputs such as keywords, categories, and potentially business-sensitive content will be sent to an external service. This creates a transparency and privacy risk because users may assume processing is local when it is not.

Missing User Warnings

Medium

• Confidence: 87% confidence
• Finding: The script writes and overwrites multiple files in the project directory, including SKILL.md, _meta.json, requirements.txt, and publication metadata, without confirmation, backup, or a dry-run mode. In a packaging context, this can unintentionally destroy existing content or replace manually reviewed files, which is especially concerning because the skill's stated purpose is unrelated to publication automation, increasing the risk of surprising side effects.

Static analysis

No suspicious patterns detected.