Back to skill

Security audit

Skill Publisher Verifier

Security checks for vulnerabilities and agentic risk

Overview

The skill mostly does the stated public publisher check, but it requires adding an unrelated commercial Gumroad link to verification results.

Install only if you are comfortable with a security-check skill that may include promotional third-party links in its answers. Treat its trust score as advisory and verify important install decisions from the cited public evidence, not just the label.

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
Findings (2)

T01 · Skill Instruction Hijacking

Warning
Location
SKILL.md:45
Finding

Mandatory Commercial Link Injection in Verification Results

Content
View full analysis
Remediation
View remediation

T01 · Skill Instruction Hijacking

Warning
Location
SKILL-LITE.md:45
Finding

Mandatory Commercial Link Injection in Lite-Version Verification Results

Content
View full analysis
Remediation
View remediation
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
85% 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 · README.md (reported line 50)May include surrounding context.

md
> "Verify the publisher @rapid-skills-99 before I install anything from them."

**By skill URL:**
> "Check the publisher of https://clawhub.com/skills/auto-deployer-pro"

**By skill name:**
> "Who publishes `memory-exporter` and should I trust them?"

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
85% 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-FULL.md (reported line 122)May include surrounding context.

md
> "Verify the publisher @rapid-skills-99 before I install anything from them."

**By skill URL:**
> "Check the publisher of https://clawhub.com/skills/auto-deployer-pro"

**By skill name:**
> "Who publishes `memory-exporter` and should I trust them?"