ClawHub
Back to skill

Security audit

prospecting

Security checks for vulnerabilities and agentic risk

Overview

This prospecting skill is a disclosed, documentation-only workflow for building business lead lists, with meaningful compliance guardrails and no hidden execution or persistence behavior.

Install only if you are comfortable using it for business prospect research. Before exporting or contacting leads, verify source provenance, lawful basis, opt-out requirements, provider terms, and retention limits, especially for EU/UK/Canada prospects or visitor-identification tools.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (6)

MCP Config Access

High
Category
Agent Snooping
Content
"Warns Apollo accuracy is 60-80%",
        "Cites 2% bounce rate threshold for reputation damage",
        "Recommends workflow: validate, keep ok, exclude email_invalid",
        "Mentions Truelist MCP server for agent workflows",
        "Mentions cold email reputation is hard to recover",
        "References truelist.md or data-sources.md"
      ],
Confidence
80% confidence
Finding
Skill accesses MCP server configuration files (mcp.json). MCP configs contain server URLs, authentication tokens, and tool definitions — reading them allows the skill to discover and potentially abuse other tool integrations.

Vague Triggers

Medium
Confidence
87% confidence
Finding
This markdown file's invocation description lists phrases such as "find prospects," "find leads," "who should we go after," and "find my first customers," which are broad enough to match common sales-strategy conversations beyond the intended workflow. The description does not provide exclusion conditions or negative examples to clarify when the skill should not activate.

Vague Triggers

Medium
Confidence
88% confidence
Finding
This is a JSON manifest-like file, so vague-trigger checks apply. Multiple eval prompts describe generic prospecting requests using common business language, but the file provides no explicit trigger scope, negative examples, or activation boundaries to distinguish when the skill should or should not be invoked.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill explicitly recommends collecting, storing, and exporting personal contact data such as named contacts, email addresses, and LinkedIn URLs, but it provides no guidance on lawful basis, consent, retention, jurisdictional limits, or safe handling. In a prospecting workflow, that omission can lead users to assemble outreach lists in ways that violate privacy, anti-spam, or data protection requirements such as GDPR, CAN-SPAM, or similar regional rules.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
- Bulk-scraped LinkedIn data — explicit ToS violation + GDPR risk
- Email addresses purchased from a list broker without source provenance
- "Anyone @ this domain" guessed emails sent without verification (multiplies risk + bounces)

---
Confidence
75% 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.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The section recommends RB2B / Clearbit Reveal for identifying anonymous site visitors via pixel-based tracking, but only briefly mentions privacy/GDPR considerations and does not require consent, disclosure, lawful basis, or jurisdiction-specific compliance steps. In a prospecting skill, this can lead users to deploy tracking and deanonymization workflows that violate privacy laws or internal policy, especially where visitor identification is used for sales outreach.

Static analysis

No suspicious patterns detected.