Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Discord Digest
v1.1.0Generate formatted digests from Discord servers using a user token. Reads messages from selected channels/threads and creates concise summaries with links. U...
⭐ 0· 569·0 current·0 all-time
byNikoxyz@nikolaybohdanov
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
medium confidencePurpose & Capability
The name/description match the implementation: the scripts use a Discord user token to read guilds, channels, threads and messages and format digests. However, the registry metadata does not declare the sensitive credential (the skill relies on a Discord user token kept in a config file rather than an explicit declared primary credential), which is a gap between manifest and actual requirements.
Instruction Scope
SKILL.md and README give concrete, narrow instructions (obtain a user token via browser devtools, run the provided Python scripts). The runtime instructions and scripts only perform Discord API reads and local formatting; they do not reference unrelated system files or external endpoints other than discord.com.
Install Mechanism
No install spec is provided (scripts run directly with Python). All code is included in the skill bundle and there are no downloads from external/untrusted URLs — low install risk.
Credentials
The skill requires a Discord user token (stored in ~/.openclaw/workspace/config/discord-digest.json by default). That token is highly privileged (full user account access) even though the code only issues read requests. The README mentions an optional DISCORD_DIGEST_CONFIG_DIR env var for overriding config location; that environment variable is not declared in registry metadata. Storing the token in plaintext on disk and asking users to copy the browser Authorization header are both sensitive practices.
Persistence & Privilege
always:false and the skill is user-invocable (normal). The skill persists the token and server list to a local config file under the user's OpenClaw workspace; it does not modify other skills or system-wide settings. Persisting a user token locally increases blast radius if the machine or workspace is shared or compromised.
Assessment
This skill is coherent and implements the documented digest functionality, but it requires you to supply a Discord "user token" (the Authorization header value copied from your browser). That token grants full account access if leaked or misused. Before installing: (1) understand this likely violates Discord ToS and use at your own risk; (2) prefer using a bot token or a separate, limited account instead of your main account token; (3) restrict filesystem permissions on the config file (~/.openclaw/workspace/config/discord-digest.json), and consider setting DISCORD_DIGEST_CONFIG_DIR to a secure location; (4) review the included scripts yourself (they only call discord.com endpoints currently) and run them in a trusted, isolated environment; (5) rotate your Discord token (change password / enable 2FA) if you ever suspect it was exposed. If you need lower risk, request a bot-based implementation that uses an explicit bot token and OAuth-based least-privilege access.Like a lobster shell, security has layers — review code before you run it.
latestvk97bfjmfsk7v36j9a9qtg9pn0s81dxxx
License
MIT-0
Free to use, modify, and redistribute. No attribution required.