nexus-brain
от mrnsmh
DevOps & CloudКратко
Intelligent SRE Bridge for AI-driven system diagnosis and recovery.
Описание
Что делает этот навык
SRE orchestrator that diagnoses service failures using an AI reasoning engine before taking any recovery action. Redacts passwords, tokens, and emails from logs before sending them for analysis. Routes infrastructure failures to automatic restart and code bugs to user alerts.
Forces log analysis before any restart, preventing blind recoveries that hide bugs rather than fix them.
Когда его использовать
- Diagnosing why a Docker container keeps restarting
- Auto-recovering a crashed pm2 process after log review
- Distinguishing a code bug from an infra failure during an outage
- Reviewing redacted logs before they reach an external AI service
- Gating recovery actions behind a manual approval step
Пример сценария
User asks: My API service is down, figure out what happened and recover if it's safe
- 1
Fetches recent logs for the affected service
- 2
Redacts sensitive fields (passwords, tokens, emails) from log content
- 3
Sends redacted logs to the configured AI reasoning service via bridge.py
- 4
Classifies the failure as infrastructure-related rather than a code bug
- 5
Restarts the service via docker or pm2, requesting /approve if configured
Service restarted after AI confirms infrastructure failure; a code bug would have triggered a user alert instead of a restart
Требования
- opencode installed and accessible in system PATH or local user directory
- Docker
Установка
clawhub install mrnsmh/nexus-brainВыполните команду CLI или вставьте промпт в Claude Code, Cursor, Codex — любой кодинг-агент, которым пользуетесь.