Daniil Oleynik
Community@lfuuu · Moscow
Daniil Oleynik maintains 57 Claude Code skills covering multi-role software delivery orchestration, documentation systems, DevOps operations, and quant research.
Agent Skills by Daniil Oleynik
Showing 57 vetted skills indexed across 1 GitHub repositories.
lfuuureviewer
Reviews diffs and pull requests against architecture and billing rules with read-only verdicts.
lfuuufrontend
Implements React TypeScript frontend pages, API clients, and UI states for AI Billing.
lfuuubackend
Guides backend development of a Node.js billing service with layered architecture and domain rules.
lfuuutodo
Manages idea and task journals in docs/todo markdown files through a draft-to-done lifecycle.
lfuuutesting
Runs Vitest unit, integration, and e2e tests and classifies failures for the responsible role.
lfuuuteamlead
Orchestrates full-cycle billing tasks by routing work across backend, frontend, testing, and reviewer roles.
lfuuutechwriter
Updates Russian-language project documentation, OpenAPI descriptions, and error message texts for AI Billing.
lfuuuresearch-agent
Runs bounded quant strategy research loops with backtesting, OOS validation, and robustness scoring.
lfuuuskill-judge
Evaluates Agent Skill quality against official specifications using eight scored dimensions.
lfuuumermaid-diagrams
Create software architecture, database, and process diagrams using Mermaid text syntax.
lfuuurepo-audit
Audits a repository for architecture, testing, security, and documentation issues with ranked findings.
lfuuuwrite-a-prd
Create a PRD through user interviews and codebase exploration, then submit it as a GitHub issue.
lfuuutriage-issue
Diagnose bug root causes and create GitHub issues with TDD-based fix plans.
lfuuudiagram
Renders Mermaid .mmd diagram sources into PNG images using mermaid-cli.
lfuuutechwriter-core
Maintains project documentation in docs/ through a fixed pipeline of edits, humanizer pass, validation, and index updates.
lfuuuprd-to-plan
Converts a PRD into a multi-phase implementation plan using vertical tracer-bullet slices.
lfuuuubiquitous-language
Extracts domain terminology from conversations into a DDD ubiquitous language glossary file.
lfuuubackend-core
Guides backend development through a fixed pipeline of analysis, tests, implementation, review, and reporting.
lfuuuagent-md-refactor
Refactors bloated agent instruction files into linked documentation using progressive disclosure.
lfuuugit-guardrails-claude-code
Configures Claude Code PreToolUse hooks that block dangerous git commands before execution.
lfuuurequest-refactor-plan
Creates a detailed refactoring plan with small commits and files it as a GitHub issue.
lfuuuimprove-codebase-architecture
Analyzes codebases to find module-deepening refactoring opportunities and creates GitHub RFC issues.
lfuuureviewer-core
Reviews code diffs against plans and returns binary APPROVED or CHANGES_REQUESTED verdicts.
lfuuusummarize
Generates structured multi-page summaries from large DOCX and Markdown documents via a multi-phase pipeline.
Frequently Asked Questions About Daniil Oleynik
FAQPage SchemaWhat tasks can I accomplish with lfuuu's skill collection?▼
You can orchestrate multi-role code delivery (backend, frontend, testing, review), generate Mermaid and C4 architecture diagrams, convert PRDs into phased plans and GitHub issues, audit repositories for technical debt, summarize large DOCX documents to PDF, and manage Kubernetes deployments, backups, and security audits.
Who are these skills designed for?▼
They target software engineers, tech leads, QA and reliability engineers, DevOps specialists, and technical writers working inside Claude Code or Hermes environments. Several skills serve specific personal projects like AI Billing MVP, PaasFabric platform, CMDB operations, and a retrogames wiki.
How do I install and run these skills?▼
Skills are SKILL.md packages placed in a Claude Code skills directory and invoked by name or trigger phrases. User-invocable skills accept argument hints (e.g., /repo-audit quick, /todo add). Reference-tier cores like backend-core are imported by project wrappers rather than called directly.
Are these skills open source and what do they cost?▼
The skills are free to use as prompt-based SKILL.md files; agent-md-refactor explicitly carries an MIT license. They require access to Claude Code or Hermes platforms, and some skills depend on external binaries like pandoc, wkhtmltopdf, mmdc, shfmt, and shellcheck.
What prerequisites and dependencies do the skills need?▼
Most skills require read/write file access; operational ones need Bash execution. Specific dependencies include git worktrees for new-branch, Docker Compose and k8s for ops-engineer, Terraform and Ansible for platform-engineer, Prisma or SQLAlchemy schemas for db-docs, and Perplexity access for web research.