Security audit

AI KEY·发布后复盘

Security checks for vulnerabilities and agentic risk

Overview

The skill reviews published-video performance and saves clearly disclosed lessons in the creator’s local workspace.

Use this with the intended AI KEY companion files and a clearly configured content vault. Expect it to update local memories and tracking records. The companion instructions and agenda configuration were not bundled, so their contents remain unverified. Share only the analytics and audience messages needed for the review.

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
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
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 (1)

The trigger "the numbers are in" is too broad for a post-publication video review skill.

Low

Category: Trigger Abuse
Confidence: 90% confidence
Finding: "The numbers are in" can refer to financial results, surveys, experiments, or other everyday reporting, not just published-video performance. Although the description explains the intended domain, the listed phrase does not require that context before activating the skill.
Content: No source excerpt is available for this finding.