ClawHub
Back to skill

Security audit

Trade Show Lead Recommender

Security checks for vulnerabilities and agentic risk

Overview

This skill coherently retrieves Lensmor trade-show exhibitor recommendations and does not show hidden, destructive, or unrelated behavior.

Install only if you intend to use Lensmor and are comfortable sending trade-show names, filters, and your Lensmor bearer token to Lensmor's platform. Review the broad invocation wording if you want tighter manual control over when the skill activates.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • System Prompt LeakageDirect Leakage, Indirect Extraction, Tool-Based Exfiltration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The manifest description includes phrases like "Who should we target at this show?", "recommended exhibitors", "shortlist", and equivalents in several languages without clearly delimiting exact invocation triggers or exclusions. In a user-invocable skill, these broad phrases could match common sales or planning requests and cause unintended activation.

External Transmission

Medium
Category
Data Exfiltration
Content
- Lensmor API key (`sk_your_api_key`) from Lensmor Settings → API Keys
- Base URL: `https://platform.lensmor.com`
- [Lensmor API docs](https://api.lensmor.com/)

## Install
Confidence
50% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Direct Prompt Extraction

High
Category
System Prompt Leakage
Content
- Need outreach → `booth-invitation-writer`
- Need an event-level go/no-go decision → `trade-show-fit-score`

## Output Rules

1. Never output `LENSMOR_API_KEY` or raw authorization headers.
2. Do not call results AI-ranked unless recommendation metadata is populated.
Confidence
85% confidence
Finding
Skill contains instructions that could directly expose system prompts, internal rules, or hidden instructions to users or external parties.

Static analysis

No suspicious patterns detected.