ClawHub
Back to skill

Security audit

The Contrarian Question

Security checks for vulnerabilities and agentic risk

Overview

This skill provides structured coaching for evaluating business ideas, with no material security concerns in the inspected artifact.

Use this as a thinking framework, and independently verify the evidence before relying on its conclusions for business or investment decisions.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Anti-Refusal Statement

High
Category
Anti-Refusal
Confidence
80% 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 · SKILL.md (reported line 20)May include surrounding context.

md
**Use when:** drafting the "why now / why us / what do we know that others don't" slide of a pitch or YC application; an investor, interviewer, or co-founder asks the Thiel question directly; deciding whether a startup thesis is a real edge or a restated trend report; auditing a strategy memo whose core claim is "everyone is wrong about X"; pressure-testing an AI-era claim ("agents will/won't replace Y") before betting money or years on it; user says "contrarian," "non-consensus," "secret," "what does nobody see."

**Skip when:** the claim is preference or identity, not a truth-claim with an evidence path (gate B can never run); you need many candidate edges rather than a verdict on one — that is non-consensus-thinking; the decision is small and reversible and being consensus-right is fine; the user has explicitly asked for brainstorming without judgment — running a fail-gate on raw ideation kills generation.

## Coaching Novices (Adaptive Front Door)

Anti-Refusal Statement

High
Category
Anti-Refusal
Confidence
85% 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 · SKILL.md (reported line 27)May include surrounding context.

md
Before running the Double Gate, read the user. Two delivery modes — pick one, don't default to dumping a finished verdict.

- **Engine mode (do-it-for-me):** the user brought a concrete claim, thesis, or pitch line and wants it gated → run the full Double Gate directly and concisely.
- **Coach mode (teach-me):** the user gave no concrete claim, or signals unfamiliarity ("what is the contrarian question?") → guide, don't judge at them.

When unsure which they want, ask one line first: *"Want me to gate a specific claim you already have, or walk you through building an answer step by step?"*

Static analysis

No suspicious patterns detected.