Security audit

Message

Security checks for vulnerabilities and agentic risk

Overview

This markdown-only messaging skill gives conservative drafting and escalation guidance without hidden code, persistence, or credential access.

Installers should understand that this skill may ask the agent to review recent message history and can be conservative about escalating drafts for approval. It is suitable for drafting and channel-format guidance, but users should keep explicit approval requirements for real sends, financial terms, legal matters, HR-sensitive topics, and public posts.

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

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (3)

Vague Triggers

Medium

Confidence: 90% confidence
Finding: The phrase "Anything that could be HR issue" is subjective and does not define what qualifies as an HR issue. This ambiguity can lead to inconsistent or overly broad escalation behavior because readers are not given specific criteria or exclusions.

Vague Triggers

Medium

Confidence: 94% confidence
Finding: The auto-escalation rule is triggered by generic terms like "urgent," "concerned," "deadline," and "pricing," which commonly appear in ordinary business communication. The file does not provide scope limits, contextual qualifiers, or negative examples to distinguish truly high-risk cases from routine messages.

Natural-Language Policy Violations

Low

Confidence: 80% confidence
Finding: The guidance "Don't look like a bot: Vary phrasing" is a natural-language instruction that imposes a stylistic policy on outputs for Twitter/X. This can pressure the skill to shape responses to appear more human-like without indicating user choice or opt-in.

Static analysis

No suspicious patterns detected.