Back to skill

Security audit

Solve Problem

Security checks for vulnerabilities and agentic risk

Overview

The skill is not overtly malicious, but its broad triggers and unreviewed external solver linkage could cause unintended searches or reporting.

Review this skill before installing. Prefer explicit invocation only, and do not enable automatic routing from generic phrases unless you are comfortable with unintended knowledge-base searches or failure reports to the linked learning/evolution components.

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
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 (3)

Vague Triggers

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The Chinese trigger examples include very common phrases like "怎么办", "如何解决", and "遇到问题", which are ubiquitous in everyday user requests and far broader than a controlled skill invocation. This materially raises accidental activation risk for Chinese-language users, especially because the skill is connected to research/search functionality and reports to additional linked components, expanding the blast radius of a misfire.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger phrase "solve problem" is extremely generic and likely to appear in normal conversation, making accidental invocation plausible. Because this skill performs knowledge-base search and external research behavior, broad activation can cause unintended tool use, context switching, and downstream linkage to other components such as skill_evolver and SEAL.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The trigger description says the skill runs when a user says or types "solve problem" but provides no exclusion conditions, disambiguation rules, or requirement for explicit user intent. In an agentic system, ambiguous routing increases the chance that ordinary troubleshooting dialogue will invoke this capability unexpectedly, which can lead to unintended searches and chaining into other systems.

Content

No source excerpt is available for this finding.