Security audit

Llm Service Manager 4cadc0

Security checks for vulnerabilities and agentic risk

Overview

The skill has no executable payload, but its declared purpose, repeated merged sections, and broad triggers are inconsistent enough that it should be reviewed before installation.

Review this skill before installing because it appears auto-merged from unrelated topics. It should be narrowed to one clear purpose and trigger set; if used for Windows service or licensing-related troubleshooting, require the agent to ask for explicit user confirmation before suggesting any system changes.

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 (8)

Description-Behavior Mismatch

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The manifest says the skill is about Obsidian math/callout-manager note organization, but the body later shifts into unrelated Windows/service troubleshooting and Autodesk Genuine Service content. This mismatch can cause the wrong skill to activate or mislead downstream agents/users into applying instructions in an unintended context, increasing the chance of unsafe or irrelevant assistance.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 86% confidence
Finding: The manifest description and trigger text are written entirely in Chinese, and the rest of the skill likewise assumes Chinese-language interaction without indicating that users may choose another language. Under the stated policy, forcing a specific language without opt-in is a natural-language policy concern unless the locale constraint is explicitly justified.
Content: No source excerpt is available for this finding.

Intent-Code Divergence

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The file contradicts itself about what was learned and what problem the skill addresses, first framing it as an Obsidian note-taking technique and later as software/service pop-up troubleshooting. Contradictory inline documentation undermines operator trust and can route a user into the wrong workflow, which is especially risky when the later content concerns system services and licensing-related software behavior.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The activation conditions are broad and ambiguous, such as triggering whenever a user 'needs related help.' Overbroad triggers can cause accidental invocation in ordinary conversations, leading an agent to inject irrelevant or potentially unsafe instructions into unrelated tasks.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 89% confidence
Finding: A generic trigger phrase like needing 'llm_service_manager related help' is likely to overlap with many normal requests. That increases false activations and raises the risk of confusing or unsafe workflow substitution, especially because the skill content is already inconsistent.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The merged section repeats the same broad trigger style, compounding the chance of unintended activation across duplicated skill fragments. Repetition of ambiguous triggers in merged content makes routing behavior unpredictable and hard to audit.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: This trigger block again uses ambiguous help-seeking language rather than precise activation criteria. In a skill that mixes unrelated topics, such ambiguity materially increases the chance that users are routed into the wrong instructions or troubleshooting path.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The final trigger block continues the same unspecific activation pattern without scope limits. Because the file contains multiple merged identities, an unrestricted trigger can cause persistent misrouting and accidental execution of irrelevant guidance.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.