Security audit

Key Takeaways

Security checks for vulnerabilities and agentic risk

Overview

This skill is a simple local key-takeaway extractor with some overstated documentation, but no evidence of hidden access, persistence, exfiltration, or destructive behavior.

Install only if you want a lightweight local helper for extracting key sentences. Do not expect full executive summaries, action-item extraction, or batch/export behavior unless the skill is improved; review outputs against the source text before relying on them.

A.I.G

Vulnerability Patterns

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
Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
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

Findings (7)

Tp4

High

Category: MCP Tool Poisoning
Confidence: 94% confidence
Finding: The declared description presents a broad summarization skill for many document types with multiple output styles and configurable formatting. The actual code only splits text into sentences, scores them using a fixed list of keywords like 'result', 'significant', and 'key finding', and returns the top matching sentences. Its docstring explicitly says 'medical documents,' indicating a narrower domain focus than declared. It does not generate true summaries, action items, executive summaries, or configurable formats; it only returns a fixed dictionary containing takeaways and a word count. This is a material description-behavior mismatch in both scope and advertised capabilities.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The manifest description says to use the skill whenever the user asks for 'summaries, bullet points, main points, highlights, or a TL;DR of any document or body of text.' These triggers are extremely broad, map to common everyday phrasing, and do not provide exclusion conditions or boundaries for when another skill should be used instead.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The usage guidance says to use the skill whenever the task needs summaries or takeaways from 'any document or body of text,' then adds generic statements about evidence tasks and fallback paths. This does not clearly define activation boundaries, accepted input types, or cases where the skill should not be invoked.
Content: No source excerpt is available for this finding.

Intent-Code Divergence

Medium

Category: Not specified by scanner
Confidence: 88% confidence
Finding: L052 states the skill should 'avoid undocumented side effects,' but later documentation explicitly describes writing exported files, batch output directories, and an errors.log file (L124-L150). This is an active contradiction between the implementation-oriented guidance and the documented behavior examples, not merely an omission.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: The skill description is broad enough to match many ordinary summarization requests, which can cause the agent to invoke this skill when a more specific or safer skill should handle the task. In an agent ecosystem, overly broad routing increases the chance of scope drift, inappropriate downstream processing, and accidental handling of sensitive content without tighter task-specific controls.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The summary 'Use when working with key takeaways' is a broad activation description that does not clearly define when the skill should or should not be invoked. It lacks specific trigger phrases, scope constraints, or exclusion conditions, which could cause unintended activation in many ordinary summarization contexts.
Content: No source excerpt is available for this finding.

Intent-Code Divergence

Low

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The manifest describes a general skill for extracting key takeaways from documents, meeting notes, articles, and other text content. The class docstring instead states it extracts key points from medical documents, which actively misrepresents the intended scope of the code relative to the documented skill purpose.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.