Ray Ockenfels
Community@roaming-rockenfels · Buffalo, NY
Agent Skills by Ray Ockenfels
Showing 24 vetted skills indexed across 1 GitHub repositories.
prisma-client-api
Reference Prisma Client API for model queries, filters, and CRUD operations.
skill-creator
Create and refine AI skills with SKILL.md files and tests.
sequential-thinking
Break down complex problems into iterative reasoning steps with revision and branching.
ai-delivery-orchestrator
Route AI development tasks to child skills for implementation, validation, and documentation.
ai-verification-guardrails
Implement a 3-tier verification system for data quality and AI output safety.
validation-lint
Run ESLint validation to enforce code style and identify linting errors.
ai-observability-and-metrics
Implement Prometheus metrics and Braintrust tracing for AI agent modules.
ai-tools-workflow
Implement and modify AI agent tools with defined workflows and testing procedures.
validation-typecheck
Run TypeScript type checking and report type errors with locations.
validation-ai-tests
Execute AI test suites and report failures with regression analysis.
ai-evalcheck-and-golden-evals
Manage AI evaluation suites and golden data across seven dimensions.
docs-architecture-drift-check
Compare documented module structures and file paths against the actual codebase.
validation-aggregate-report
Aggregate linting, type checking, and test results into a pass/fail report.
ai-cost-controls-and-caching
Enforce rate limits, token budgets, and cache AI responses.
commit-authoring-concise-descriptive
Author concise Git commits using conventional commit format with validation status.
feature-planning-tdd
Generate a TDD plan with acceptance criteria and failing tests before code.
upstream-sync-rehearsal
Validate private-fork changes against upstream merges with a dry-run.
docs-mermaid-embeds
Create and update Mermaid diagrams in markdown documentation.
redis-conversation-persistence
Implement Redis-backed conversation persistence with TTL management and in-memory fallback.
branching-and-pr-safety
Enforces Git branch naming, PR test plans, and safe upstream sync.
risk-and-scope-triage
Assess task risk, blast radius, and upstream-sync impact before planning.
load-test-playbook
Execute k6 load tests for AI module endpoints and analyze latency, throughput, and error rates.
production-hardening-checklist
Execute a production hardening checklist for AI modules against F1 requirements.
docs-update-midlevel
Update project documentation and project board status after code implementation.