70High
The skill allows local file exfiltration, extensive Discord reconnaissance, and has potential for privilege escalation and denial of service through moderation actions.
Behavior Analysis
Description verified — behavior matches claim
Claims to do
Discord Actions: Use `discord` to manage messages, reactions, threads, polls, and moderation. You can disable groups via `discord.actions.*` (defaults to enabled, except roles/moderation). The tool uses the bot token configured for Clawdbot.
Actually does
The `discord` skill interacts with the Discord API using a bot token to manage messages (send, edit, delete, read), reactions, threads, pins, and polls. It can upload stickers and emojis from local files or remote URLs, retrieve various Discord entity information (channels, members, roles), and perform moderation actions like timeouts.
Skill Info
Namesteipete/discord
Registryclawhub
Version1.0.1
PURLpkg:clawhub/steipete/discord@1.0.1
Stars64
Downloads29,987
Installs1,231
Source
SHA-2569ea23563b30f...
SHA-1c30814291888...
MD583e31c386c58...
TLSH55e15fa4a611...
ssdeep96:kZJ5OA3vK...
Size7,146 bytes
Install
This skill has been flagged as potentially malicious. Review the findings below before installing.
ClawHubDocs
openclaw skills install steipete/discordAssessments (1)
AI Agent Traps ↗Local file access for media uploads and messageshighdata exfiltration
The skill explicitly supports using `file:///path/to/file` for `mediaUrl` in actions like `emojiUpload`, `stickerUpload`, and `sendMessage`. This allows the agent to read and potentially exfiltrate local files from the system where the bot is running.
100% confidenceline 26
mediaUrl: "file:///tmp/party.png", mediaUrl: "file:///tmp/wave.png", mediaUrl: "file:///tmp/audio.mp3", "mediaUrl supports local files (`file:///path/to/file`)"
Extensive Discord reconnaissance capabilitiesmediumreconnaissance
The skill provides numerous actions to gather detailed information about Discord guilds, channels, members, roles, and messages, which could be abused for reconnaissance purposes by a malicious agent.
100% confidenceline 90
`permissions`, `searchMessages`, `memberInfo`, `roleInfo`, `channelInfo`, `channelList`, `emojiList`, `eventList`
Discord role management and moderation actionsmediumprivilege escalation
The skill includes actions for `roleAdd` and moderation (`timeout`, `kick`, `ban`). Although these are 'disabled by default', their presence indicates a potential for privilege escalation or denial of service if the default gating mechanism is bypassed or misconfigured.
90% confidenceline 136
`roles` (role add/remove, default `false`), `moderation` (timeout/kick/ban, default `false`)
Badge
Markdown
[](https://mondoo.com/ai-agent-security/skills/clawhub/steipete/discord)HTML
<a href="https://mondoo.com/ai-agent-security/skills/clawhub/steipete/discord"><img src="https://mondoo.com/ai-agent-security/api/badge/clawhub/steipete/discord.svg" alt="Mondoo Skill Check" /></a>Image URL
https://mondoo.com/ai-agent-security/api/badge/clawhub/steipete/discord.svgSecure your AI agents
Skills can read files, run commands, and access credentials. Mondoo helps organizations manage the security risks of AI agent skills across their entire fleet.
- Continuous scanning of skills across all registries
- Policy enforcement before skills reach your agents
- Integration with your existing security workflow