Claude Skills by Perun-Engineering

github.com/Perun-Engineering

15 skillsA× 150 installs0 views

Capture SnapshotACapture a snapshot of the top 10 CloudWatch Log groups by ingestion volume in the last 15 minutes, plus an error-line count for each via a bounded Logs Insights query.devopsawsVotes: 0GitHub stars: 3
Discover Log GroupsAList the account's real CloudWatch log groups and return those matching alert keywords (ranked, capped). Run this first so queries target log groups that actually exist instead of guessed names.devopsVotes: 0GitHub stars: 3
Query Cloudwatch LogsAExecute a CloudWatch Logs Insights query against log groups returned by discover_log_groups within an investigation window. Validates group existence, skips missing ones, returns structured findings.devopsVotes: 0GitHub stars: 3
Capture SnapshotACapture a read-only snapshot of Discord bot reachability — bot identity and guild/channel access count.devopsVotes: 0GitHub stars: 3
Scan Discord ChannelsAScan up to 10 Discord channels for alerts, notifications, and error messages from bots within a configurable investigation window.devopsVotes: 0GitHub stars: 3
Capture SnapshotACapture a read-only snapshot of EKS cluster state — node readiness, pod phases across all namespaces, non-Running pods, and recent kube-system warning events.devopsnodekubernetesVotes: 0GitHub stars: 3
Detect Recent ChangesAChange-correlation for EKS — answer "what changed right before the alert?" by reading each implicated Deployment's rollout state (spec vs observed generation, Progressing condition), container image tags, and retained ReplicaSet history. Reports a ranked lead like "image v1→v2 at 12:20, ~10 min before the alert". Read-only.devopskubernetesapiVotes: 0GitHub stars: 3
Gather Eks StateAGather Kubernetes cluster state for specified resources in a namespace — pod status, recent events, pod descriptions, pod logs (tail), node conditions. Handles unreachable EKS API server gracefully.devopsnodekubernetesVotes: 0GitHub stars: 3
Capture SnapshotAReport how many incident-outcome records have been written to the incident-history store in the last 30 days.devopsVotes: 0GitHub stars: 3
Find Similar IncidentsAFind past investigations whose alert text is similar to the current alert, ranked by embedding cosine similarity, with their recorded root cause and a link back to the original incident thread.devopsVotes: 0GitHub stars: 3
Capture Status SnapshotACapture a read-only status snapshot of all infrastructure observed by the SRE agents and post it to chat. Used for the operator-driven `/sre-snapshot` command.devopsVotes: 0GitHub stars: 3
Finalize PostmortemAAssemble and post a Post-Incident Report (PIR) for an incident thread. Used for the operator-driven `/postmortem` command.devopsVotes: 0GitHub stars: 3
Investigate AlertAReceive an AlertContext, orchestrate parallel investigation across enabled agents, synthesize the Incident Report within 60 seconds, and post enrichment updates until the 5-minute cutoff.devopsVotes: 0GitHub stars: 3
Capture SnapshotACapture a read-only snapshot of Slack workspace reachability — bot authentication and channel-membership count.devopsVotes: 0GitHub stars: 3
Scan Slack ChannelsAScan up to 10 Slack channels for alerts, notifications, and error messages from integration bots within a configurable investigation window.devopsVotes: 0GitHub stars: 3