How does this skill detect a Cohere outage?

It utilizes curl commands to check the official status.cohere.com API and performs direct endpoint tests on Chat, Embed, and Rerank services to verify real-time functionality.

Does it support both trial and production Cohere keys?

Yes, it includes specific logic to identify trial key limitations (429 errors) and provides workflows for rotating keys or upgrading to production credentials.

What happens if the Cohere API is completely down?

The skill provides 'Graceful Degradation' patterns and code snippets to implement circuit breakers, fallback models, or cached response strategies until the service is restored.

Can this skill fix Kubernetes issues related to Cohere?

Yes, it includes pre-defined commands for secret management, environment variable updates, and pod restarts via kubectl to resolve local integration issues.

Cohere Incident Runbook

Name: Cohere Incident Runbook
Author: jeremylongshore

byjeremylongshore

•

2,028

•

分析与监控

Provides automated procedures and standardized protocols for managing Cohere API service outages and performance degradations.

The Cohere Incident Runbook skill equips Claude with specialized knowledge and tools to handle service interruptions within Cohere's AI infrastructure. It streamlines the incident response lifecycle—from initial triage and endpoint health checks using status.cohere.com APIs to implementing immediate mitigations like API key rotation or circuit-breaking logic. By providing standardized communication templates and post-incident review structures, it ensures that developers can maintain service reliability and clear stakeholder communication during high-pressure system failures.

主要功能

01Built-in communication templates for internal Slack and external status updates

02Infrastructure-level mitigation commands for Kubernetes and local environments

03Evidence collection and automated postmortem generation tools

04Automated triage for Cohere Chat, Embed, and Rerank endpoints

05Standardized severity level mapping and decision trees for rapid response

062,028 GitHub stars

使用场景

01Responding to sudden 5xx errors or global API outages from Cohere

02Managing rate-limiting (429) issues by switching from trial to production keys

03Performing post-incident analysis to improve system resilience and fallback logic

主要功能

01Built-in communication templates for internal Slack and external status updates

02Infrastructure-level mitigation commands for Kubernetes and local environments

03Evidence collection and automated postmortem generation tools

04Automated triage for Cohere Chat, Embed, and Rerank endpoints

05Standardized severity level mapping and decision trees for rapid response

062,028 GitHub stars

使用场景

01Responding to sudden 5xx errors or global API outages from Cohere

02Managing rate-limiting (429) issues by switching from trial to production keys

03Performing post-incident analysis to improve system resilience and fallback logic