From discord-remote
Set up discord-remote — save the bot token, configure approval settings, review sidecar status, and manage deny patterns. Use when the user pastes a Discord bot token, asks to configure discord-remote, asks about remote approval setup, or wants to check sidecar status.
How this skill is triggered — by the user, by Claude, or both
Slash command
/discord-remote:configureThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Sets up and configures the discord-remote plugin. Writes the bot token to
Sets up and configures the discord-remote plugin. Writes the bot token to
~/.claude/channels/discord/.env and manages remote-config.json for
approval settings, timeouts, and deny patterns.
Arguments passed: $ARGUMENTS
Read state files and give the user a complete picture:
Token — check ~/.claude/channels/discord/.env for
DISCORD_BOT_TOKEN. Show set/not-set; if set, show first 6 chars masked.
Sidecar — check if ~/.claude/channels/discord/sidecar.port exists.
If it does, read the port and try curl -s http://127.0.0.1:<port>/health
to confirm the sidecar is alive. Show: running/not-running, port, pending
request count if running.
Remote config — read ~/.claude/channels/discord/remote-config.json
(missing = defaults). Show:
Access — read ~/.claude/channels/discord/access.json (missing =
defaults). Show: dmPolicy, allowFrom count (this is the user who receives
approval DMs).
What next — end with a concrete next step based on state:
"Run /discord-remote:configure <token> with your bot token.""DM your bot on Discord. It replies with a code; approve with /discord-remote:access pair <code>.""The MCP server is not running. Start a Claude Code session to activate it.""Ready. Tool approval requests will be forwarded to Discord."<token> — save bot token$ARGUMENTS as the token (trim whitespace).mkdir -p ~/.claude/channels/discord.env if present; update/add the DISCORD_BOT_TOKEN= line,
preserve other keys. Write back, no quotes around the value.chmod 600 ~/.claude/channels/discord/.env — the token is a credential.timeout <approval_ms> [question_ms] — configure timeoutsremote-config.json (create with defaults if missing).timeout.approval_ms to the first argument.timeout.question_ms.fallback <ask|allow|deny> — configure permission fallbackask, allow, deny.remote-config.json, update defaults.permission_fallback, write back.ask (default): falls back to terminal prompt — nothing breaks if
Discord is unreachableallow: automatically allows if no Discord response — use only if you
trust all tool invocationsdeny: automatically denies if no Discord response — safe but blocks
progress if Discord is downdeny add <pattern> — add a deny patternremote-config.json.deny_patterns (dedupe).deny rm <pattern> — remove a deny patternremote-config.json.deny_patterns.deny list — list deny patternsremote-config.json. Show all deny patterns, numbered.clear — remove the bot tokenDelete the DISCORD_BOT_TOKEN= line (or the file if that's the only line).
If the official discord plugin is also installed, both plugins share the
same bot token and access.json. This is intentional — they share access
control so you don't have to pair twice. The two plugins will send separate
DMs for their respective operations (channel messaging vs. approval requests),
which may appear duplicated if both are active simultaneously.
The discord-remote plugin uses a different sidecar port file
(sidecar.port) that the official plugin does not write or read.
.env once at boot. Token changes need a session restart
or /reload-plugins. Say so after saving.remote-config.json is read on each sidecar request (Windows) or at
startup + SIGHUP (Unix). Changes take effect on next request after SIGHUP,
or immediately on Windows. Say so after editing.npx claudepluginhub lizardliang/lizard-market --plugin discord-remoteWrites Discord bot token to config and guides you through channel setup, access policy, and allowlist lockdown.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.