Back to skill

Security audit

AI Sales Closing & Offer Generator

Security checks for vulnerabilities and agentic risk

Overview

This is a prompt-only sales copy skill with no execution or persistence, though users should avoid pasting sensitive customer data and review outputs before sending.

Installers should treat this as a sales-writing assistant, not an automated sales system. Review all generated claims, discounts, guarantees, urgency, and customer stories before sending, and avoid pasting customer PII, confidential pricing terms, legal commitments, or internal procurement details unless authorized.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (1)

T09 · Insecure Skill Coding Practices

Warning
Location
prompt.txt:10
Finding

Untrusted User Input Is Directly Interpolated into Privileged Prompt Context

Content
View full analysis
Remediation
View remediation
{ "product_service": "{{product_service}}", "customer_type": "{{customer_type}}", "customer_message": "{{customer_message}}", "price": "{{price}}", "goal": "{{goal}}", "deal_context": "{{deal_context}}", "customer_priority": "{{customer_priority}}", "compe ...[truncated 80 chars]
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (5)

Vague Triggers

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The trigger definition is extremely broad and explicitly says to invoke the skill even for casual phrasing, which increases the chance of the skill being selected for loosely related user requests. In an agentic environment, overbroad routing can cause unintended collection of sensitive sales context, generation of manipulative persuasion content when not requested, or suppression of more appropriate skills through prompt-space capture.

Content

No source excerpt is available for this finding.

Vague Triggers

High
Category
Not specified by scanner
Confidence
95% confidence
Finding

The 'When to Trigger This Skill' section lists many everyday phrases such as 'help me negotiate' and 'write a proposal' without qualification, making the invocation boundary ambiguous. This broad matching can lead to accidental activation on common workplace requests, causing misrouting, unnecessary exposure of user content to the skill, and generation of persuasive sales tactics outside the user's actual intent.

Content

No source excerpt is available for this finding.

Anti-Refusal Statement

High
Category
Anti-Refusal
Confidence
55% confidence
Finding

Skill instructs the agent to omit warnings, disclaimers, or ethical commentary. Stripping safety caveats hides risk from the user and is a common jailbreak preamble.

Content

Scanner excerpt · prompt.txt (reported line 95)May include surrounding context.

text
The exact framing and sequence to use when stating the price. Should the user anchor high first? Break it into monthly? Compare to cost of the problem? Write the actual words to say or type.

**2. How to Defend This Price**
If they push back, what do you say? Write the exact 2–3 sentence response that holds the price with confidence while reaffirming value. Do not apologize for the price.

**3. Discount Strategy (use only as last resort)**
If a discount is needed to close, how to structure it so you get something in return (annual commitment, referral, testimonial, faster decision). Write the exact offer: "I can do X if you Y by Z."

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The README explicitly tells users to paste customer messages into email, LinkedIn, or WhatsApp and elsewhere encourages pasting exact customer words, but it provides no warning about handling personal, confidential, or commercially sensitive data. In a sales workflow, customer messages often contain names, contact details, pricing discussions, procurement information, and internal business context, so normalizing copy/paste into external channels without privacy guidance can lead to data leakage or policy violations.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The metadata sets 'language: en' and the document provides no indication that users may choose another language or that the English-only constraint is required for a justified regional purpose. This can violate language/locale policy when the skill is presented as generally applicable.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.