tkerseyCommunity·29 Agent Skills Included

dotfiles

Rigorous code review, proof, planning, and PR landing workflows

Orchestrates disciplined software work: counterexample-driven repair, invariant design, PR review reconciliation, safe merging, and proof-backed publishing. Eliminates vague reviews, unverified claims, stale evidence, and ceremonial process that never changes code. Includes specialized skills for Lean 4 proofs, Zig engineering, performance optimization, planning, and persistent engineering memory.

66|2

↗

npx skills add tkersey/dotfiles --all -g -y

Available:

Instructs the agent to prioritize direct working capability over process ceremony, route requests to the correct sub-skills, handle concurrent edits safely, and follow strict Git, Python, and ledger tooling rules.

All Skills in This Repository (29)

Pure Emerald Level Indicators

📦 In Repo

tkersey

mesh

Delegate inline task lists to subagents with dependency-aware parallel execution.

commit

Create small Git commits with validation signals before finalizing.

zig

Guide Zig 0.15.2 development with runbook, fuzzing, and performance templates.

ms

Create, update, and refactor Codex skills with SKILL.md and agents/openai.yaml.

fin

Finish GitHub pull requests by updating branches, monitoring CI, and squash-merging.

creative-problem-solver

Generate a five-tier CPS portfolio with signals and escape hatches.

join

Automate GitHub PR management with CI gates and squash-merge via gh CLI.

gen-plan

Generate sequential plan-N.md files in the repository root.

reduce

Audits codebase abstractions and proposes prioritized cuts with a migration plan.

slice

Generate or validate a dependency-aware DAG of work slices in SLICES.md.

seq

Parse Codex JSONL session logs to extract skill usage and token counts.

tk

Enforce contracts and invariants to turn change requests into verifiable patches.

Community

Advanced

Frequently Asked Questions

FAQPage Schema

How to install tkersey dotfiles skills?▼

Run `npx skills add tkersey/dotfiles --all -g -y` in your terminal to install all skills in this collection globally.

What problems do these skills solve?▼

They enforce evidence-based engineering: reviews must cite real counterexamples, merges require resolved threads and live state proof, and performance claims require before-and-after measurements.

Can these skills help merge a pull request safely?▼

Yes. The land skill reconciles every unresolved review thread, verifies merge admission against the exact head, proves the live merged state, and cleans up branches and worktrees.

Do these skills work with Codex and Claude Code?▼

Yes. All skills follow the standard SKILL.md format and run in compatible coding environments including Codex, Claude Code, and similar runtimes.

Do I need formal methods experience to use them?▼

No. You describe the task in plain English and the agent applies the right workflow, whether that is a Lean proof, a Zig audit, or a code review.