opsphere-io/opsphere

CI and pipeline failure investigator for GitHub Actions, Bitbucket, and GitLab. Use when workflows or pipelines fail.

Help users configure how Opsphere knows their deployment sources — work context (Community) vs admin Cloud Catalog (Team). Use when deployment_status returns gaps, user asks about deployment sources, or latest-deploy queries fail for missing catalog config.

Connect an eligible provider (Datadog, Vercel, GitHub, Cloudflare, Jira, Confluence, Sentry, Bitbucket, GitLab, Railway, …) when ops_list_integrations marks it available_to_connect — step-by-step guided setup from the chat. AWS is excluded (Cloud Catalog admin only). Use when the user wants to add, configure, or reconnect a provider that is configure_cta-eligible, or when a tool reports missing credentials for an eligible module.

Connect the same Opsphere account to another MCP app without sharing tokens or changing workspace.

Single-host endpoint health — DNS, HTTP, and TLS. Use when the user asks if a URL is up, SSL expiry, or DNS resolution.

Site-down and incident triage. Use when the user reports outages, downtime, 5xx errors, degraded SLOs, or widespread failures.

Get started with Opsphere in Antigravity; discover tools, account and workspace without changing configuration.

Structured incident post-mortem writer. Use after an outage when the user wants RCA summary or to save lessons learned.

Assess whether a release or current deployment for a named environment and optional site, product, service, or workspace is ready to promote using matching evidence.

Design focused QA cases and investigate suspected product defects with Opsphere evidence. Use for feature validation, reproduction, regression checks, and deciding whether a reported behavior is a bug.

Read or update Opsphere provider and account context from Antigravity; not workspace selection or credential setup.