Discord Project Manager

• Outcome: Pass
• Latest audit: May 28, 2026 at 9:49 AM UTC
• Version:
  2.1.6

Overview

This skill is powerful but coherent: it manages Discord/OpenClaw project permissions as advertised and discloses its sensitive config and token access.

Install only if you want this skill to manage Discord project channels and OpenClaw agent permissions. Use a least-privilege Discord bot limited to the intended guild, back up ~/.openclaw/openclaw.json before first use, and double-check archive/remove commands because they can revoke agent access and trigger a gateway reload.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• 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

Findings (5)

Context-Inappropriate Capability

Medium

• Confidence: 95% confidence
• Finding: The skill reads a Discord bot token from the user's local OpenClaw configuration, giving it access to a sensitive credential without explicit user consent or scoping. In an agent-skill context, this expands the skill's privilege boundary and could enable unauthorized Discord API actions if the skill is misused or modified.

Context-Inappropriate Capability

Medium

• Confidence: 91% confidence
• Finding: The skill reads a global user-scoped OpenClaw configuration file from ~/.openclaw/openclaw.json and derives Discord guild information from it without any explicit consent, scoping check, or validation that the data is intended for this skill. This creates a cross-boundary information access issue: the skill can consume unrelated application configuration and potentially bind itself to the wrong Discord environment if the shared config is stale, maliciously modified, or contains multiple accounts/guilds.

Missing User Warnings

Medium

• Confidence: 86% confidence
• Finding: The fallback path directly edits a sensitive configuration file and sends a signal to a gateway process without any confirmation, authorization check, or explicit operator warning in this layer. In an agent skill context that manages Discord permissions, this increases the risk of unintended or unauthorized reconfiguration affecting service behavior or access controls.

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: The code silently retrieves a bot token from a local config file, with no user-facing disclosure that a stored credential is being accessed. In agent environments, undisclosed secret access is risky because users may not realize the skill can act with their bot's privileges across Discord resources.

Missing User Warnings

Medium

• Confidence: 88% confidence
• Finding: The archive operation is a destructive action that removes permissions for all registered agents and marks the project archived immediately, with no confirmation, dry-run, or secondary authorization step. In a multi-agent Discord coordination system, a mistaken or spoofed invocation could lock participants out of a thread and disrupt collaboration state, making accidental misuse materially harmful.

VirusTotal

66/66 vendors flagged this skill as clean.