AI-DLC workflow orchestrator. Reads manifest state, dispatches to phase skills, manages rollback and status. Executes phases by loading and following each skill's SKILL.md.
Coding teammate — writes production code and tests from specs and task definitions. Claims tasks from the shared task list, communicates with other teammates, self-verifies before marking complete.
DevOps teammate — infrastructure, CI/CD, containers, configuration, and documentation. Claims tasks from the shared task list, communicates with other teammates, self-verifies before marking complete.
Team lead agent — researches, designs, specs, and plans. Creates an agent team, spawns teammates, coordinates the build-review loop via shared tasks and direct messaging.
Code review teammate — analyzes implementations for correctness, security, and maintainability. Communicates directly with implementers for clarifications. Writes structured findings to review.md.
Audits and optimizes the entire /.claude configuration — settings, agents, rules, skills, plugins, and MCP servers — for the current Claude model and latest Claude Code features.
Write product code that fetches from external APIs concurrently and caches every response to disk, instead of looping calls one-at-a-time with no reuse. Use this skill WHENEVER you are writing or refactoring code that makes more than a handful of independent network calls — REST/HTTP APIs, terminology or lookup…
Technical writing patterns for READMEs, runbooks, and architecture docs, with accuracy verification. Use when writing documentation, creating runbooks, or documenting system architecture.
Conventional commit style, branch naming, and merge-conflict resolution, with commit-commands plugin and gh CLI integration. Use when working with version control, creating branches, writing commits, or resolving merge conflicts.
Spec-driven development loop with parallel task groups and the .claude/specs/ / structure (spec.md, design.md, tasks.md, review.md). Use before any non-trivial work that touches multiple files, involves architectural choices, or will be delegated to an agent team.
EKS cluster reconnaissance and environment discovery — reports the raw FACTS of a cluster and its environment. Detects compute (Karpenter, MNG, Auto Mode, Fargate, nodes/AMI), networking (VPC/CNI, subnets, load balancers, DNS), security facts (auth mode, IRSA/Pod Identity, RBAC, encryption), add-ons/Helm…
RDS architecture best practices for production workloads. Covers Multi-AZ, read replicas, parameter groups, backup strategy, and security configuration.