Cloud and SaaS Outage Triage: check vendors before changing code

[koko3tallah]

The Cloud and SaaS Outage Triage agent is now merged into Awesome Copilot and live in the searchable catalog.

The failure mode it addresses is simple: an application starts timing out, an agent sees errors in the repository, and debugging begins before anyone checks whether AWS, Cloudflare, GitHub, OpenAI, Anthropic, Stripe, or another upstream dependency is already reporting an incident.

Install the approved agent from its catalog page:

https://awesome-copilot.github.com/agent/cloud-saas-outage-triage/

Useful prompts to try:

• Before debugging this deploy, check AWS, Cloudflare, GitHub, and OpenAI.
• Is Claude down, or is this failure in my app?
• Build a timestamped health snapshot for the external services used by this repository.
• Correlate this error window with any active provider incident, then keep investigating locally if the evidence does not match.

The agent uses nine read-only OutageDeck MCP tools to search providers, check a stack, inspect current status and incident timelines, and review uptime history. Public checks are free and require no account or API key. The workflow deliberately treats official status as one signal rather than proof: it matches product, region, symptom, and timing, cites source links, and continues local investigation when the provider evidence is stale or incomplete.

The MCP server currently normalizes official feeds for 172 cloud and SaaS providers. Setup and tool details are at https://outagedeck.com/developers/mcp?utm_source=github&utm_medium=discussion&utm_campaign=awesome_copilot_showcase

Disclosure: OutageDeck maintains both the agent contribution and the hosted MCP service. The public status tools used by this agent are read-only, free without signup, and the agent excludes account-scoped mutation tools.