ClawHub
Back to skill

Security audit

Iran Briefing

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed read-only briefing helper that queries Capduck public intelligence feeds and does not request local access, credentials, persistence, or mutation authority.

Install only if you are comfortable with the agent querying Capduck's remote feeds for geopolitical, Truth Social, news, and prediction-market information. Ask the agent to answer in your preferred language if you do not want bilingual output.

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
  • System Prompt LeakageDirect Leakage, Indirect Extraction, Tool-Based Exfiltration
  • 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 (2)

Direct Prompt Extraction

High
Category
System Prompt Leakage
Confidence
85% confidence
Finding

Skill contains instructions that could directly expose system prompts, internal rules, or hidden instructions to users or external parties.

Content

Scanner excerpt · SKILL.md (reported line 205)May include surrounding context.

md
**"Deep dive on a specific event"** — Get event ID from briefing or `/events` -> call `/posts?event={id}` and `/news?event={id}` -> triangulate source count and source types.

## Output guidelines

- **Lead with judgment, then evidence.** Don't dump raw data — synthesize first, cite after.
- **Mark confidence.** Use "confirmed (N sources)" / "reported but unverified (single source)" / "rumor" based on `source_count`.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The notes state that all responses are Markdown with Chinese translations where available, and the Trump feed is bilingual by default. This creates a language/locale behavior that is imposed by the skill rather than offered as a user choice, which can violate language preference policy absent explicit opt-in or justification.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.