Back to skill

Security audit

News Summarizer Official

Security checks for vulnerabilities and agentic risk

Overview

This skill fetches public news feeds and can optionally turn a generated news summary into audio using OpenAI TTS, with no hidden persistence or unrelated access found.

Installers should understand that voice summaries require sending the generated summary text to OpenAI and creating /tmp/news.mp3 locally. Use the text-only workflow for sensitive or proprietary prompts, and delete the temporary audio file if it should not remain on disk.

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
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill instructs sending generated summary text to an external TTS API and writing the resulting audio to /tmp/news.mp3 without any notice, consent step, or handling guidance for potentially sensitive content. Even if the expected input is public news, user prompts or synthesized summaries may still include private or proprietary information, so silent transmission and local file creation create privacy and data-handling risks.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
86% confidence
Finding

The referenced API endpoint is an external network destination, so using it causes data egress from the agent environment. While the destination itself is expected for TTS, the lack of warning, consent, and data-classification guidance makes this a real privacy and compliance concern rather than a harmless implementation detail.

Content

Scanner excerpt · SKILL.md (reported line 82)May include surrounding context.

  1. Send as audio message
bash
curl -s https://api.openai.com/v1/audio/speech \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{

External Transmission

Medium
Category
Data Exfiltration
Confidence
86% confidence
Finding

The referenced API endpoint is an external network destination, so using it causes data egress from the agent environment. While the destination itself is expected for TTS, the lack of warning, consent, and data-classification guidance makes this a real privacy and compliance concern rather than a harmless implementation detail.

Content

Scanner excerpt · SKILL.md (reported line 82)May include surrounding context.

  1. Send as audio message
bash
curl -s https://api.openai.com/v1/audio/speech \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{