Security audit

doc-coauthoring

Security checks for vulnerabilities and agentic risk

Overview

This skill is labeled for documentation co-authoring but asks for broad automation and command execution, so users should review it carefully before installing.

Install only if you are comfortable granting a documentation-labeled skill command execution authority. Prefer a version that limits itself to document drafting and editing, or that clearly lists exactly what commands and files it may use.

SkillSpector

Vulnerability Patterns

Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
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

Findings (5)

Description-Behavior Mismatch

High

Confidence: 96% confidence
Finding: The skill is presented as a narrow documentation co-authoring workflow, but the body expands it into generic automation, data processing, API usage, file handling, and command execution. This capability mismatch can mislead users and policy layers into granting broader authority than expected, increasing the chance of unsafe execution or abuse under a benign-looking label.

Context-Inappropriate Capability

High

Confidence: 98% confidence
Finding: Requesting exec capability for a documentation co-authoring skill is unjustified on its face and materially increases risk because command execution can access files, invoke external tools, or alter the environment. In a skill whose stated purpose is writing docs, this creates an unnecessary privilege escalation path if the agent or user invokes the skill expecting only text assistance.

Intent-Code Divergence

Medium

Confidence: 88% confidence
Finding: The documentation inconsistently describes the skill as both document co-authoring and broad Development automation/data processing. This ambiguity weakens user understanding and review controls, making it easier for risky behavior to be normalized or triggered outside the intended context.

Vague Triggers

Medium

Confidence: 85% confidence
Finding: The trigger condition 'when the user needs to execute related operations' is so broad that it provides little practical boundary for activation. Overly permissive activation criteria increase the likelihood that the skill is invoked in unintended contexts, especially dangerous here because the skill advertises read and exec capabilities.

Vague Triggers

Medium

Confidence: 80% confidence
Finding: The description uses broad activation language spanning Development automation, data analysis, and orchestration without clear scope limits. In combination with privileged tools, this makes the skill easier to invoke for unrelated tasks and obscures what behavior reviewers and users should expect.

Static analysis

No suspicious patterns detected.