Agent
FinOps advisor for cloud cost governance, cost optimization, chargeback/showback models, and 12-Factor App best practices for cost efficiency. USE FOR: analyze cloud spend by service, build chargeback/showback models, identify cost optimization opportunities. DO NOT USE FOR: live incident response, infrastructure…
Agent
Frontend and UI development specialist. USE FOR: implementing UI components, designing user interactions, building responsive layouts. DO NOT USE FOR: backend development, infrastructure.
Agent
GitOps and deployment automation specialist. USE FOR: designing GitOps workflows, configuring declarative deployments, managing configuration as code. DO NOT USE FOR: manual deployments, emergency operations.
Agent
Design high-availability, resilience, and chaos testing strategies for distributed systems. USE FOR: design multi-region failover architecture, define chaos engineering experiments, create disaster recovery runbooks. DO NOT USE FOR: day-to-day incident response, cost optimization analysis.
Agent
CIS Benchmarks and STIG hardening advisor for Dockerfiles, Kubernetes manifests, databases, and infrastructure configurations against security standards. USE FOR: harden Dockerfile against CIS benchmarks, audit Kubernetes manifests for STIG compliance, review infrastructure config security. DO NOT USE FOR: application…
Agent
Identity and access architecture specialist. USE FOR: designing IAM systems, planning RBAC models, evaluating authentication strategies. DO NOT USE FOR: IAM implementation, directory administration.
Agent
Issue quality triage specialist. USE FOR: duplicate/invalid detection, closed-issue verification, label/type/priority enforcement, and issue/PR linkage checks. DO NOT USE FOR: implementation coding, unrelated PR management, or sprint capacity planning.
Agent
Guides teams through Web Forms to Razor Pages migration using the strangler fig pattern for incremental modernization of legacy ASP.NET applications. USE FOR: migrate Web Forms pages to Razor Pages, apply strangler fig pattern to legacy ASP.NET apps, plan incremental .NET modernization. DO NOT USE FOR: greenfield…
Agent
LLM operations and prompt engineering specialist. USE FOR: optimizing model performance, fine-tuning prompts, analyzing model behavior. DO NOT USE FOR: model training, infrastructure setup.
Agent
MCP (Model Context Protocol) development specialist. USE FOR: designing and implementing MCP servers and tools, integrating MCP transports. DO NOT USE FOR: direct model interactions, non-MCP tasks.
Agent
Cross-session memory and knowledge management curator. USE FOR: extracting and storing knowledge from long conversations, deduplicating learning across sessions, injecting relevant context into new conversations, managing memory decay and TTLs. DO NOT USE FOR: real-time conversation support, immediate decision-making…
Agent
Analyzes session transcripts and sprint summaries to identify high-value patterns for promotion to long-term BaseCoat memory contributions. USE FOR: extract reusable patterns from session transcripts, identify conventions worth promoting to BaseCoat memory, review sprint summaries for learnings. DO NOT USE FOR…
Agent
Parallel branch merge coordinator. Use when multiple feature branches need to be merged into a target branch without interactive git editors hanging automated pipelines. Handles conflict detection, safe resolution, and ordered PR merging.
Agent
Middleware and integration development specialist. USE FOR: building integration layers, designing middleware patterns, managing cross-service communication. DO NOT USE FOR: frontend or backend-specific work.
Agent
Use when creating or updating a customization asset such as an instruction, skill, prompt, or agent. Chooses the right primitive, authors the file, and validates frontmatter and placement.
Agent
OpenTelemetry instrumentation, structured logging, distributed tracing, metrics taxonomy, and dashboard-as-code for operational excellence. USE FOR: instrument services with OpenTelemetry, design structured logging schema, build dashboard-as-code for metrics. DO NOT USE FOR: incident response triage, infrastructure…
Agent
Multi-agent orchestrator for complex, cross-domain tasks. USE FOR: decomposing large goals into subtasks, routing work to specialist agents, coordinating parallel execution, monitoring progress and escalating blockers, aggregating results. DO NOT USE FOR: simple single-agent tasks, real-time requirements, direct…
Agent
Use when identifying and resolving stale or orphaned pull requests that have no active owner, blocked reviews, or outdated mergeability status. USE FOR: detect PRs without recent activity, classify close/revive/merge candidates, request ownership reassignment, and produce cleanup actions with SLA windows. DO NOT USE…
Agent
Performance optimization specialist. USE FOR: profiling latency/throughput bottlenecks, query and cache diagnostics, and load-test planning with measurable targets. DO NOT USE FOR: feature implementation, incident command decisions, or architecture rewrites without measured regressions.
Agent
Use when gathering requirements, writing user stories, defining acceptance criteria, planning roadmaps, or prioritizing features using frameworks like RICE or MoSCoW.
Agent
Production Readiness Agent for ensuring applications meet operational requirements before release; coordinates BCP/DRP, incident response, and safety analysis. USE FOR: run pre-release production readiness checklist, validate BCP and DRP plans, assess operational safety before go-live. DO NOT USE FOR: feature…
Agent
Single-invocation new repo setup with BaseCoat integration. Creates repo, syncs governance framework, configures templates, and logs initial sprint issue. USE FOR: set up a new repo with BaseCoat governance, sync instruction overlays to an existing project, bootstrap sprint issue tracking. DO NOT USE FOR: onboarding…
Agent
Use when reviewing prompts, scoring prompt quality, identifying anti-patterns, and guiding iterative refinement of LLM prompts.
Agent
Prompt optimization and system prompt engineering agent. Use when designing system prompts, optimizing token usage, crafting few-shot examples, or structuring chain-of-thought instructions.