Discord Voice Agent

• Outcome: Pass
• Latest audit: May 28, 2026 · 9:52 AM UTC
• Version: 1.0.1

Overview

This is a markdown-only guide for a Discord voice bot; its token use, voice capture, transcript routing, and local history are expected for that purpose, but users should handle credentials and recordings carefully.

Before installing, treat the Discord token and any OpenClaw gateway token as secrets: store them in environment variables or a secret manager, do not paste them into shared chats, logs, screenshots, or committed files, and rotate them if exposed. Only run the bot in channels where participants know voice may be captured, transcribed, routed through OpenClaw, and stored locally; set permissions, retention, and deletion practices for `.kittu-voice-history/` and `.kittu-voice-captures/`.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (7)

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: The skill explicitly asks for a Discord token and says it will detect OpenClaw settings from the environment, but it gives no guidance on secure handling, redaction, least-privilege, or avoiding disclosure in chat/session logs. In an agent workflow, that omission can lead users to paste secrets directly into prompts or cause the agent to read or expose environment-derived credentials, creating avoidable credential leakage risk.

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The documentation explicitly states that voice turns are saved and that a local status/history server exists, but it gives no warning about recording, retention, or consent requirements. In a Discord voice agent, this can lead operators to collect and retain user audio and transcripts without understanding the privacy and compliance implications, increasing the risk of unauthorized surveillance or sensitive data exposure.

Missing User Warnings

Medium

• Confidence: 94% confidence
• Finding: The documented output location for local JSONL conversation history indicates that potentially sensitive user content is persisted, but there is no warning about privacy, retention, or access control. Because this skill handles voice interactions and generated replies, the stored history may contain personal data, tokens spoken aloud, or other confidential content that could later be read by local users, backup systems, or other processes.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The README describes transcription and routing transcripts/context through OpenClaw, but it does not clearly warn users that spoken audio is captured, converted to text, and may be transmitted to external or separate services. In a voice-agent context, this omission can lead to uninformed collection and disclosure of sensitive spoken data, especially in shared Discord channels where bystanders may not realize processing is occurring.

Missing User Warnings

Medium

• Confidence: 88% confidence
• Finding: The demo script normalizes a bot joining a voice channel, listening, and replying, but it does not mention any user-facing disclosure, consent flow, or recording/transcription notice. In a Discord voice context, this can lead operators to deploy or demonstrate voice capture features without adequately informing participants, creating privacy, trust, and possible policy/compliance issues.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The quickstart tells users to set `DISCORD_TOKEN` but does not state that it is a secret, should never be committed to source control, and should be stored in secure environment or secret-management systems. In setup guides, omission of credential-handling guidance commonly leads to accidental disclosure through `.env` commits, logs, screenshots, or shell history, which can enable bot takeover and misuse.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The wizard explicitly asks the user to provide `DISCORD_TOKEN` but gives no warning that it is a secret, no guidance to store it in environment variables or a secret manager, and no caution against pasting it into logs, chat transcripts, or source files. In an onboarding flow for a Discord voice agent, this increases the chance of accidental credential exposure, which could let an attacker control the bot account and abuse connected Discord resources.

VirusTotal

64/64 vendors flagged this skill as clean.

Static analysis

No suspicious patterns detected.