Back to skill

Security audit

Local Stt Workflow

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent local speech-to-text debugging skill, with privacy caveats around audio files and log/proxy captures but no evidence of hidden or unsafe behavior.

Install only if you are debugging a local STT workflow. Use non-sensitive sample audio where possible, be careful with gateway logs or proxy captures because they may contain transcripts and metadata, and remove temporary captures after debugging.

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 (2)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

This is a markdown file, so SQP-2 applies to omissions in user-facing documentation. The document instructs users to upload audio files and optionally provide contextual text, but it does not warn that these inputs may contain sensitive personal or confidential information and will be processed by the service.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

The skill recommends using a dump proxy or server logs to inspect transcription traffic but does not warn that those requests can contain sensitive audio content, transcripts, filenames, and metadata. In a local STT debugging workflow this is common operational guidance, but without minimization or redaction guidance it can lead to unnecessary exposure of private user data during troubleshooting.

Content

No source excerpt is available for this finding.