mod-actions

Moderator Actions

Take moderator actions on Civitai users including banning, muting, removing content, managing leaderboard eligibility, and sending direct messages.

This skill uses the Civitai tRPC API with API key authentication, ensuring all actions go through the proper service layer with full side effects (session invalidation, search index updates, activity tracking, etc.).

Setup

1. Copy .env-example to .env in this skill directory
2. Add your Civitai API key (must belong to a moderator account)
3. Optionally set the API URL (defaults to production)

cp .claude/skills/mod-actions/.env.example .claude/skills/mod-actions/.env
# Edit .env and add your API key

Get your API key from: https://civitai.com/user/account (API Keys section)

Important: The API key must belong to a user with moderator privileges.

Environment Variables

Variable	Required	Default	Description
CIVITAI_API_KEY	Yes	-	Your Civitai API key
CIVITAI_API_URL	No	https://civitai.com	API base URL

For local development, set CIVITAI_API_URL=http://localhost:3000

Running Commands

node .claude/skills/mod-actions/query.mjs <command> [options]

Commands

Command	Description
user <id|username>	Look up user info by ID or username
ban <id|username>	Ban a user (toggle - will unban if already banned)
mute <id|username>	Mute a user (toggle - will unmute if already muted)
leaderboard <id|username> <true|false>	Set leaderboard eligibility
remove-content <id|username>	Remove all content from a user
dm <id|username>	Send a DM to a user (requires --message)

Ban Reason Codes

Code	Description
SexualMinor	Sexual content involving minors
SexualMinorGenerator	Generator for sexual minor content
SexualMinorTraining	Training on sexual minor content
SexualPOI	Sexual content with people of interest
Bestiality	Bestiality content
Scat	Scat content
Nudify	Nudification tools
Harassment	Harassment
LeaderboardCheating	Leaderboard manipulation
BuzzCheating	Buzz system abuse
RRDViolation	Rights and Restrictions Denial violation
Other	Other violation

Options

Flag	Description
--json	Output raw JSON response
--dry-run	Show what would be done without making changes
--reason <code>	Ban reason code (for ban command)
--message <text>	External message shown to user (for ban command)
--internal <text>	Internal notes (for ban command)

Examples

Look Up User

# By user ID
node .claude/skills/mod-actions/query.mjs user 3879899

# By username
node .claude/skills/mod-actions/query.mjs user unfazedanomaly964

# Get raw JSON
node .claude/skills/mod-actions/query.mjs user 3879899 --json

Ban User

# Simple ban (toggles ban status)
node .claude/skills/mod-actions/query.mjs ban 3879899

# Ban with reason and message
node .claude/skills/mod-actions/query.mjs ban 3879899 --reason Other --message "Repeated ToS violations"

# Ban with internal notes
node .claude/skills/mod-actions/query.mjs ban 3879899 --reason Other --message "ToS violation" --internal "User exploited republish bug 15 times"

# Ban by username
node .claude/skills/mod-actions/query.mjs ban unfazedanomaly964 --reason Other

# Dry run to see what would happen
node .claude/skills/mod-actions/query.mjs ban 3879899 --dry-run

Mute User

# Toggle mute status
node .claude/skills/mod-actions/query.mjs mute 3879899

# By username
node .claude/skills/mod-actions/query.mjs mute unfazedanomaly964

# Dry run
node .claude/skills/mod-actions/query.mjs mute 3879899 --dry-run

Manage Leaderboard Eligibility

# Exclude from leaderboards
node .claude/skills/mod-actions/query.mjs leaderboard 3879899 false

# Include in leaderboards
node .claude/skills/mod-actions/query.mjs leaderboard 3879899 true

# Dry run
node .claude/skills/mod-actions/query.mjs leaderboard 3879899 false --dry-run

Remove All Content

# Remove all content from a user (DESTRUCTIVE)
node .claude/skills/mod-actions/query.mjs remove-content 3879899

# Always dry run first!
node .claude/skills/mod-actions/query.mjs remove-content 3879899 --dry-run

Output Format

User Info

User: unfazedanomaly964
ID: 3879899
Status: Active
Banned: No
Muted: No
Leaderboard Eligible: Yes
Created: 2024-03-19

Action Result

Action: BAN
User: unfazedanomaly964 (ID: 3879899)
Success: Yes
Previous: Not Banned
Now: Banned
Reason: Other

How It Works

This skill calls the Civitai tRPC API endpoints:

• user.getById / user.getCreator - User lookups
• user.toggleBan - Ban/unban users
• user.toggleMute - Mute/unmute users
• user.setLeaderboardEligibility - Manage leaderboard access
• user.removeAllContent - Remove all user content

Authentication is via Bearer token in the Authorization header:

Authorization: Bearer <your-api-key>

All endpoints require moderatorProcedure access, meaning the API key must belong to a moderator account.

When to Use

• Investigating violations: Look up user info before taking action
• Repeat offenders: Ban users who repeatedly violate ToS
• Temporary restrictions: Mute users for minor violations
• Leaderboard manipulation: Exclude cheaters from leaderboards
• Content violations: Remove all content from severe violators

Safety Notes

• Always use --dry-run first for destructive actions
• All actions are logged in the ModActivity table
• Ban actions invalidate user sessions immediately
• Universal bans unpublish all user content and cancel subscriptions
• Document reasons with --reason and --message flags
• API key must have moderator privileges

Tips

• Always look up user info first to confirm the right user
• Use --dry-run for destructive actions like bans and content removal
• Document reasons with --reason, --message, and --internal flags
• Use --json for scripting or piping to other tools
• Set CIVITAI_API_URL=http://localhost:3000 for local development