ClawHub
Back to skill

Security audit

RFC Writer

Security checks for vulnerabilities and agentic risk

Overview

This skill is a text-only RFC drafting helper with no executable code, data access, persistence, or hidden high-impact behavior.

Before installing, expect this skill to format qualifying technical proposal requests as RFCs and to mirror Chinese or English based on the prompt. No special security permissions are evident, but users who want a lighter response style may need to ask explicitly outside the RFC format.

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
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill description contains broad trigger phrases like 'write an RFC,' 'draft a technical proposal,' and 'structure an architecture design,' which can overlap with many normal user requests. This can cause the skill to activate in situations where the user did not explicitly intend to invoke it, potentially steering responses into a rigid template or interfering with more appropriate specialized skills.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
90% confidence
Finding

The line explicitly says the skill supports bilingual output and will 'automatically adapt' to the user's question language. Under the policy, forcing or automatically selecting a language/locale without an explicit user choice or opt-in can be a natural-language policy issue.

Content

No source excerpt is available for this finding.

Vague Triggers

Low
Category
Not specified by scanner
Confidence
77% confidence
Finding

This manifest-like JSON shows the skill handling both Chinese and English RFC-writing requests, but it provides no explicit trigger scope, invocation constraints, or exclusion conditions. For manifest/config-style files, the absence of specific activation boundaries can make it unclear when the skill should be selected versus other general writing skills.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
81% confidence
Finding

The expected outputs require one RFC in Chinese and another in English, indicating the skill may enforce language based on the input examples, but the file does not document any user language choice policy or opt-in behavior. Because locale/language policy applies to all file types, this can be a mild policy concern if the skill defaults users into a language without explicit selection rules.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.