tkogut
Community@tkogut
tkogut maintains an agentic SDLC skill registry covering autonomous PR pipelines, issue triage, spec writing, UX review, and LLM cost observability.
Agent Skills by tkogut
Showing 77 vetted skills indexed across 1 GitHub repositories.
tkogutom-discover
Guides product discovery interviews and writes an evidence-tagged product-brief.md.
tkogutom-merge-buddy
Classify open pull requests by merge readiness from reviews, CI checks, and labels.
tkogutom-setup-agent-pipeline
Configures the shared agentic pipeline config, tracker descriptors, and project docs for a repository.
tkogutom-spec-writing
Write and review feature specifications with skeleton-first drafting and severity-ranked architectural review.
tkogutom-auto-fix-issue
Automates end-to-end tracker issue resolution from triage through PR review and QA.
tkogutom-auto-create-pr
Executes autonomous coding tasks end-to-end and ships them as labeled pull requests.
tkogutom-ux-setup
Extracts a repository's design tokens, components, and screen archetypes into a committed .uxproof/ contract.
tkogutn8n-ops
Manage n8n workflows on a Hostinger VPS via docker exec import, export, and backup operations.
tkogutom-synthetic-users
Generates evidence-grounded persona panels that interview and walk product flows as synthetic users.
tkogutom-fix
Implements minimal code fixes from analyzer briefs with regression tests and validation gates.
tkogutom-auto-implement-spec
Implements an existing spec into a reviewed, UI-verified pull request.
tkogutom-approve-merge-pr
Approves and squash-merges a pull request after enforcing QA gates and label blocks.
tkogutom-auto-qa-pr
Verifies a PR's UI changes in a real browser and produces screenshot evidence with a pass/fail report.
tkogutom-pipeline-retro
Classify finished pull request pipeline runs and rank rework causes by wall-clock hours.
tkogutskill-rebuild
Synchronizes the local agent environment with the latest Vault templates and global skills.
tkogutswarm-onboarding
Generates an executive onboarding briefing from Git state, MEMORY.md, task.md, and architectural plans.
tkogutom-brainstorm
Guides divergent pre-spec conversations that route ideas to the correct next pipeline skill.
tkogutlogic-auditor
Scans source files for logic, architecture, and consistency errors before deployment.
tkoguttest-creator
Generates unit, integration, and E2E test templates for Vitest, Pytest, and Playwright.
tkogutom-auto-review-pr
Reviews pull requests in isolated worktrees and runs autonomous autofix loops until merge-ready.
tkogutom-ux-shape
Turn vague product and AI feature ideas into decided, implementation-ready directions.
tkogutgithub-orchestrator
Automates GitHub repository creation, commits, workflow runs, and pull requests via the gh CLI.
tkogutom-root-cause
Diagnoses tracker issue root causes and defines minimal code change sets.
tkogutom-close-fixed-issues
Close tracker issues that merged pull requests authoritatively fixed and comment on unresolved ones.
Frequently Asked Questions About tkogut
FAQPage SchemaWhat tasks can I automate using tkogut's skills?▼
You can run autonomous PR pipelines (create, continue, review, fix, QA, approve-merge), triage and fix tracker issues end-to-end, write staff-level specs, build backlogs from product briefs, run browser-based UI QA with screenshots, and monitor or reduce LLM token spend via the Caveman Cloud skill family.
Who are these skills designed for?▼
Software engineers, tech leads, and product-minded developers running agent-assisted development in IDEs like Claude Code, Cursor, or Antigravity. The om-* pipeline suits teams enforcing SDLC discipline; caveman-* skills target operators optimizing LLM costs; UX skills serve product designers validating flows.
How do I install and configure the pipeline?▼
Run om-setup-agent-pipeline once per repository: it inspects the default branch, validation scripts, and labels, writes .ai/agentic.config.json, installs the tracker descriptor, and generates SDLC.md, CODE_REVIEW.md, and AGENTS.md. Re-run it whenever the toolchain or label taxonomy changes.
What prerequisites do the skills require?▼
A git repository with a configured issue tracker, validation scripts for the quality gate, and optionally a browser provider for UI QA and screenshot evidence. VPS and n8n skills require Docker Compose and Traefik on a Hostinger VPS; Caveman skills require wiring the repo through the Caveman Cloud gateway.
Do the skills support resuming long-running implementations?▼
Yes. om-auto-create-pr-loop handles multi-step spec implementations with a run folder (PLAN/HANDOFF/NOTIFY), per-step commits, and checkpoint verification every ~5 steps. Interrupted runs resume via om-auto-continue-pr or om-auto-continue-pr-loop from the first unchecked plan step in an isolated worktree.