Social Post Generator

• Outcome: Pass
• Latest audit: Sep 11, 2026 · 2:35 AM UTC
• Version: 1.2.0

Overview

This is a low-risk social media drafting helper with a few documentation and routing quality issues but no hidden posting, data access, or persistence.

Install this as a drafting aid, not an autoposter. Review generated social posts before publishing, and be aware that broad trigger phrases may activate it for generic social-writing requests. The publisher should clarify language behavior and fix the hashtag script documentation so it does not imply current internet or API access.

A.I.G

Vulnerability Patterns

• Skill Instruction Hijacking: Alters the agent's session goals or safety constraints when the skill loads
• Agent Memory Poisoning: Writes attacker-controlled rules into memory that affect later sessions
• Remote Payload Retrieval and Execution: Fetches external code whose behavior can change after review
• Embedded Malicious Code: Ships malicious scripts inside the skill and executes them locally
• Unauthorized Access and Privilege Escalation: Obtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• YARA Signatures: Malware Match, Webshell Match, Cryptominer Match
• 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 (9)

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 94% confidence
• Finding: The declared description presents a broad social media content generation skill covering post creation for multiple platforms, tone control, repurposing, and formatting. The supplied code chunk only implements a narrow subset: basic hashtag generation from a topic. Although hashtag generation is explicitly mentioned in the description, the actual code lacks the primary advertised capabilities and therefore materially underdelivers relative to the declared purpose. There is no evidence of undeclared sensitive behavior; the imported urllib/json modules are unused, so there is no actual internet access in this chunk despite the comment.

Vague Triggers

Medium

• Confidence: 93% confidence
• Finding: The README advertises very generic trigger phrases such as "Write a tweet about [topic]" and "Help me with hashtags for [topic]," which are likely to overlap with ordinary user requests unrelated to explicitly invoking this skill. In a skill-routing system, broad triggers can cause unintended activation, leading to misrouting, unexpected behavior, or accidental exposure of the skill's instructions over more appropriate handlers.

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 70% confidence
• Finding: Without declared permissions the skill's intent is opaque and cannot be validated.

Vague Triggers

Medium

• Confidence: 93% confidence
• Finding: The trigger phrases are broad enough that the skill could activate for many generic requests about writing or posting, potentially routing user content into this skill unexpectedly. Over-broad activation increases the chance of unintended behavior, prompt collisions with other skills, and processing of sensitive or irrelevant user input without clear user intent.

YARA rule 'network_reconnaissance': Network reconnaissance and scanning patterns [hacktools]

Medium

• Category: YARA Match

Content

day.

I was wrong.

After years of trial and error, I discovered 6 time management principles that actually work:

1️⃣ The 2-Minute Rule
If something takes less than 2 minutes, do it immediately. This prevents small tasks from piling up into mental clutter.

2️⃣ Time Blocking
Your calendar should reflect your priorities, not other people's agendas. Block time for deep work and protect it fiercely.

3️⃣ The Power of "No"
Every commitment is a trade-off. Before saying yes, ask: "Will this move the needle?"

4️⃣ Energy Management
Work with your natural rhythms. Schedule demanding tasks during your peak energy hours.

5️⃣ Single-Tasking
Multitasking reduces productivity by up to 40%. Focus on one thing at a time.

6️⃣ Weekly Reviews
Every week, reflect on what worked, what didn't, and what to improve.

The result?

• 3x more output
• Less stress
• Better work-life balance

Productivity isn't about doing more.
It's about doing what matters.

What's your #1 tim

• Confidence: 65% confidence
• Finding: YARA rule matched a hack tool or exploit indicator (offensive tools, reconnaissance, privilege escalation, or exploit frameworks).

Natural-Language Policy Violations

Low

• Confidence: 88% confidence
• Finding: The branding and creator-facing text includes Chinese-language content, but the document does not indicate whether users can choose their preferred language or that the Chinese content is optional. This can create a language/locale policy concern if the skill is expected to avoid imposing a language without user opt-in.

Intent-Code Divergence

Low

• Confidence: 93% confidence
• Finding: The module documentation says the script 'requires internet access to fetch trending data' and refers to offline alternatives, but the implemented code never makes any network request and generates hashtags entirely from local string manipulation and hardcoded mappings. This is an active contradiction between documentation and actual behavior, not just missing detail.

Intent-Code Divergence

Low

• Confidence: 75% confidence
• Finding: The function docstring states that in production it would call Twitter/Instagram APIs, but the current function only returns locally generated hashtag combinations. While partly clarifying future intent, this documentation can mislead readers about what the code actually does today.

Natural-Language Policy Violations

Low

• Confidence: 84% confidence
• Finding: This markdown test report presents headings, instructions, and evaluations in Chinese while the generated outputs are largely in English. Because the file does not state that bilingual or English output is intentional or user-selected, it may violate a language/locale policy requiring user opt-in rather than forcing a language choice implicitly.

Static analysis

No suspicious patterns detected.