ClawHub
Back to skill

Security audit

Clarity Submit

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward Clarity Protocol submission/status tool, but users should only run it when they intend to send research details to that external service.

Install only if you are comfortable sending submitted protein, variant, rationale, optional disease, and optional wallet information to Clarity Protocol. Keep CLARITY_WRITE_KEY scoped and private, and ask the agent to use this skill only for explicit submissions or status checks.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (5)

Tp4

High
Category
MCP Tool Poisoning
Confidence
91% confidence
Finding
The primary purpose mostly matches: this script submits a protein variant hypothesis to Clarity Protocol. It also aligns with proposing variants, queuing investigation metadata, and returning a tracking URL consistent with follow-up processing. However, the declared capabilities include 'check submission status,' and no such functionality exists in this code chunk. The script only performs submission and displays the initial response. The mention of CLARITY_WRITE_KEY is not contradicted by the code because authentication could be handled inside api_post, but it is not evident here. Therefore this is a partial description-behavior mismatch due to overclaiming a capability not present in the supplied code.

Lp3

Medium
Category
MCP Least Privilege
Confidence
92% confidence
Finding
The skill declares network and environment-variable dependent behavior but does not define explicit tool scope or permissions boundaries. That omission can cause an agent framework to invoke the skill without clear user-consent or least-privilege constraints, increasing the risk of unintended outbound requests and use of CLARITY_WRITE_KEY.

Vague Triggers

Medium
Confidence
95% confidence
Finding
The invocation guidance uses broad triggers like proposing variants, investigating mutations, or queueing folds without tight scope boundaries or confirmation requirements. In an agent setting, that can cause over-triggering and unintentional submission of user-supplied research content to an external service, especially when a write key is present.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill does not clearly warn that user-provided protein, variant, rationale, and optional wallet data are sent to a third-party service. This weakens informed consent and can expose sensitive unpublished research or identifying information through external transmission and retention.

External Transmission

Medium
Category
Data Exfiltration
Content
url = API_BASE + endpoint

    try:
        response = requests.post(
            url,
            json=data,
            headers=get_headers(write=True),
Confidence
80% 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.