ClawHub
Back to skill

Security audit

Dialogue Next Steps

Security checks for vulnerabilities and agentic risk

Overview

This is a simple conversation-formatting skill that guides answers and optional next-step suggestions without requesting system access or hidden behavior.

Install this if you want the assistant to add structured follow-up options for beginner, conceptual, or underspecified questions. The only practical caution is style-related: it may make responses longer by adding next steps unless the user asks for no suggestions.

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 (3)

Anti-Refusal Statement

High
Category
Anti-Refusal
Content
### Output format

- **Always answer first**.
- If gated on, append a localized heading and a numbered list.

Example skeleton:
Confidence
86% confidence
Finding
The instruction to 'Always answer first' can weaken safety behavior by biasing the agent toward compliance before it evaluates whether refusal, de-escalation, or a safety warning is required. In isolation this is not overtly malicious, but in harmful or policy-restricted contexts it can suppress safer response patterns and increase the chance of answering disallowed requests.

Anti-Refusal Statement

High
Category
Anti-Refusal
Content
## Output Format (when `add_next_steps = true`)

- Always answer first.
- Then append a localized heading and a numbered list (up to 5). Use the user's language.

Template:
Confidence
70% confidence
Finding
Skill instructs the agent to never refuse or to always comply. Suppressing the agent's ability to decline removes a core safety control and enables downstream harmful requests to succeed.

Natural-Language Policy Violations

Low
Confidence
85% confidence
Finding
The README presents the skill instructions in English and Chinese only via a fixed language split, which can imply a constrained locale/language policy rather than user choice. There is no explicit statement that the skill should adapt to the user's preferred language or ask for language preference.

Static analysis

No suspicious patterns detected.