From cortex
Tail or follow cortex service logs from Docker Compose. Use when the user asks for cortex service logs, startup logs, crash logs, plugin deployment logs, Docker logs, or follow mode. This is for the service's stdout/stderr, not client syslog entries.
How this skill is triggered — by the user, by Claude, or both
Slash command
/cortex:cortex-logsThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Show recent cortex service logs. These are the binary's stdout/stderr, not syslog entries received from clients. Use the `cortex` MCP tool with `action=tail` for received log entries.
Show recent cortex service logs. These are the binary's stdout/stderr, not syslog entries received from clients. Use the cortex MCP tool with action=tail for received log entries.
Check server mode:
echo "$CLAUDE_PLUGIN_OPTION_IS_SERVER"
If false, explain that the local plugin is in client mode and has no local service logs to tail.
Parse user arguments:
100: that many lines, no follow--follow or -f: last 50 lines, then streamRun Docker Compose logs:
CORTEX_HOME="${CORTEX_HOME:-$HOME/.cortex}"
docker compose --env-file "${CORTEX_HOME}/.env" -f "${CORTEX_HOME}/compose/docker-compose.yml" logs cortex --tail <N> --no-color
For follow mode, add -f and omit --no-color.
If the plugin compose project has no container, report that Docker mode is configured but no plugin-managed container is running. Do not guess a source checkout path.
Show the captured logs directly. If follow mode is active, the streaming output is the deliverable until the user interrupts.
If the logs show obvious unhealthy patterns such as repeated restarts, panics, bind errors, or DB lock errors, summarize the pattern after the log excerpt and suggest cortex-dr for full diagnostics.
npx claudepluginhub jmagar/dendrite --plugin cortexCreates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.