Back to skill

Security audit

Rankizz Cloud Site Audit

Security checks for vulnerabilities and agentic risk

Overview

The skill was rated benign with high confidence: scans completed across all four components, and the only finding was a low-probability excessive-agency alert judged not expected for the skill’s stated purpose.

The scanners report a benign result with high confidence and complete coverage of the four listed components. Consider whether the skill’s stated purpose fits your needs before installing; a benign verdict does not guarantee zero risk.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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 (2)

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
80% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 16)May include surrounding context.

md
## Resolve access safely

Prefer Rankizz MCP in interactive AI clients. Use the REST API only for headless clients or when the user explicitly chooses it. Never ask the user to paste an API key into chat; direct them to `https://www.rankizz.com/api-access` and their client's secret storage.

Use OAuth for ChatGPT, Claude.ai, and other interactive remote MCP clients. The MCP endpoint is:

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
80% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 51)May include surrounding context.

md
## Enrich only when it changes the decision

Do not automatically run unrelated paid research.

- Use Search Console or Bing Webmaster only when the user asks about indexing or connected search performance.
- Use Google Analytics only when organic landing-page behavior or conversions affect prioritization.