Security audit

info-collector

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed public-information gathering skill with no executable installer or hidden code, but users should control scope, language, and any saved findings.

Before installing, decide whether you are comfortable with a skill that searches public news, forums, repositories, and social platforms for a topic. Specify the language, time window, included/excluded sources, and avoid targeting private individuals or sensitive communities unless you have a lawful and appropriate reason. Do not save findings to memory unless you want those summaries retained for later use.

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

Findings (6)

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 89% confidence
Finding: The skill explicitly supports collecting and analyzing data from social media, forums, and public opinion sources, but the description does not warn users about privacy, profiling, misinformation, or jurisdiction-specific compliance risks. This omission can cause users to employ the skill for monitoring individuals or communities without informed consent or awareness of legal and ethical constraints.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: Setting zh-CN as the default output language without explicit user opt-in can create consent and transparency issues, especially if users expect neutral locale handling or if language choice affects source selection, analysis framing, or downstream review. In an information-gathering skill, silent defaults may bias collection toward particular regions or make outputs less understandable to users, increasing the chance of misuse or missed verification.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The scheduled-task scenario is underspecified and does not define authorization, scope limits, frequency, or topic constraints. That ambiguity can let the skill run automatically in contexts the user did not clearly approve, leading to repeated network activity, unnecessary data collection, or policy-violating automation.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 88% confidence
Finding: The scheduled-task scenario is underspecified and does not define authorization, scope limits, frequency, or topic constraints. That ambiguity can let the skill run automatically in contexts the user did not clearly approve, leading to repeated network activity, unnecessary data collection, or policy-violating automation.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 97% confidence
Finding: Defaulting output language to zh-CN without user opt-in can produce responses in a language the user did not request, which is a scope and UX safety issue. While not directly enabling code execution or data exfiltration, it can degrade comprehension, cause mistaken reliance on untranslated summaries, and create unexpected behavior in multilingual environments.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The execution template repeats and operationalizes the fixed zh-CN default, increasing the likelihood that downstream invocations will enforce a language the user did not choose. This makes the issue more persistent because it is embedded in the run instructions, not just the metadata.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.