ClawHub
Back to skill

Security audit

GauntletScore

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed cloud API integration for checking documents and code, with no local execution or hidden install behavior found.

Install only if you are comfortable sending the documents, code, or URLs you choose to analyze to GauntletScore's cloud API. Do not submit secrets, private repositories, internal URLs, regulated data, or client material unless you have approval and have reviewed the vendor's privacy and retention terms; use the Sovereign Edition for no-egress requirements.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (7)

External Script Fetching

High
Category
Supply Chain
Content
### In code:
- Reverse shells and remote code execution
- Credential theft and data exfiltration
- Download-and-execute attacks (curl | bash)
- Prompt manipulation attempts (instructions designed to override agent behavior)
- Documentation-vs-behavior mismatches (code does things the README doesn't mention)
- Dangerous operation combinations that pass individual checks
Confidence
90% confidence
Finding
Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill clearly instructs users to submit document text, code, or a skill URL to a third-party cloud API, but it does not prominently warn that potentially sensitive content leaves the local environment for remote processing. This can cause unintended disclosure of proprietary code, internal documents, or regulated data because users may assume a local-only verification workflow from the security-focused marketing language.

External Transmission

Medium
Category
Data Exfiltration
Content
### Verify a document by pasting content:
```
POST https://api.gauntletscore.com/v1/analyze
Authorization: Bearer YOUR_API_KEY
Content-Type: application/json
Confidence
96% confidence
Finding
The skill instructs users to POST raw document contents to an external API endpoint. External transmission is expected for a SaaS verifier, but it is still a real security/privacy risk because sensitive material may be exfiltrated from the user's environment to a third party without sufficiently prominent disclosure or safeguards in the skill text.

External Transmission

Medium
Category
Data Exfiltration
Content
### Verify a ClawHub skill by URL:
```
POST https://api.gauntletscore.com/v1/analyze
Authorization: Bearer YOUR_API_KEY
Content-Type: application/json
Confidence
94% confidence
Finding
The skill encourages sending a source_url to an external API for analysis, which may expose internal or access-controlled locations if users provide non-public URLs. In a security-validation context, users may be more likely to trust the workflow and submit sensitive targets, increasing the chance of unintended disclosure to the vendor.

Skill Enumeration

Medium
Category
Agent Snooping
Content
Content-Type: application/json

{
  "source_url": "https://clawhub.ai/skills/gauntlet-validate/SKILL.md",
  "topic": "Evaluate the safety of this ClawHub skill before installation"
}
```
Confidence
80% confidence
Finding
Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

External Transmission

Medium
Category
Data Exfiltration
Content
### Check results:
```
GET https://api.gauntletscore.com/v1/jobs/{job_id}
Authorization: Bearer YOUR_API_KEY
```
Confidence
50% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

External Transmission

Medium
Category
Data Exfiltration
Content
### Check results:
```
GET https://api.gauntletscore.com/v1/jobs/{job_id}
Authorization: Bearer YOUR_API_KEY
```
Confidence
50% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Static analysis

No suspicious patterns detected.