ai-devops-toolkitOperational tooling for teams running local LLM infrastructure. Request tracing with full scoring breakdowns, per-application usage analytics via request tag...
Install via ClawdBot CLI:
clawdbot install twinsgeeks/ai-devops-toolkitGrade Fair — based on market validation, documentation quality, package completeness, maintenance status, and authenticity signals.
Sends data to undocumented external endpoint (potential exfiltration)
POST → http://localhost:11435/dashboard/api/settingsCalls external URL not in known-safe list
https://github.com/geeks-accelerator/ollama-herdAI Analysis
The skill interacts with a local router (localhost:11435) for observability of a user's own LLM infrastructure, consistent with its stated DevOps purpose. The external GitHub URL is for documentation and package sourcing, not data exfiltration. No evidence of credential harvesting, hidden instructions, or obfuscation was found.
Audited Apr 17, 2026 · audit v1.0
Generated May 23, 2026
A site reliability engineer uses the toolkit to monitor latency percentiles, error rates, and node health across a fleet of Ollama instances. They can quickly identify degraded nodes, investigate failed requests with full trace data, and trigger alerts based on severity thresholds without any external monitoring stack.
A product manager needs to track token usage and request volumes per application or team. By leveraging the request tagging feature and SQLite analytics, they can generate per-app usage reports and allocate infrastructure costs based on actual consumption patterns.
An infrastructure lead uses the model recommendation and usage analytics to forecast when additional GPUs or nodes are needed. The automated health checks and underutilized node detection help optimize resource allocation and avoid over-provisioning.
A DevOps team integrates the toolkit's health check API into their incident response pipeline. When critical checks fail (e.g., offline nodes or high error rates), automated remediation can restart nodes or reroute traffic, reducing mean time to recovery.
A managed LLM service provider uses the toolkit to offer per-tenant dashboards showing latency, failures, and usage. The SQLite backend keeps the observability layer self-contained and cost-effective for each tenant, with no additional infrastructure required.
The toolkit is open source, but the company offers paid support, SLAs, and enterprise features like advanced alerting or integration with existing observability tools. Revenue comes from annual subscription fees for support and premium features.
Offer a hosted version of the observability layer where customers upload their SQLite databases or connect their fleet for centralized monitoring. Revenue is based on the number of requests traced or nodes monitored per month.
Provide consulting to help teams deploy and customize the toolkit for their specific LLM infrastructure. Revenue comes from one-time setup fees and ongoing optimization contracts, leveraging expertise in local LLM operations.
💬 Integration Tip
To integrate, ensure the Ollama Herd router is running at localhost:11435 and install sqlite3 for direct database queries. Use the Python example scripts to quickly generate reports or integrate with existing monitoring dashboards via the REST API.
Scored May 23, 2026
Automatically update Clawdbot and all installed skills once daily. Runs via cron, checks for updates, applies them, and messages the user with a summary of what changed.
Analyze Twitter content to uncover WordPress and Shopify client pain points, craft authority-building posts, and generate qualified inbound lead insights.
Real-time security monitoring for Clawdbot. Detects intrusions, unusual API calls, credential usage patterns, and alerts on breaches.
Interact with Uptime Kuma monitoring server. Use for checking monitor status, adding/removing monitors, pausing/resuming checks, viewing heartbeat history. Triggers on mentions of Uptime Kuma, server monitoring, uptime checks, or service health monitoring.
A clean, reliable system resource monitor for CPU load, RAM, Swap, and Disk usage. Optimized for OpenClaw.
Publish content to Mastodon. Use when you need to post a Mastodon status.