Pedro Nauck
Community@pedronauck · Florianópolis
Building harness for AI
Agent Skills by Pedro Nauck
Showing 242 vetted skills indexed across 1 GitHub repositories.
refactoring-analysis
Audits codebases for Fowler code smells and writes prioritized refactoring reports.
ship-pr
Automates the end-of-feature workflow of generating release notes, writing PR descriptions, committing, and opening GitHub PRs.
qa-execution
Runs persona-driven dogfooding QA sessions through a product's public interfaces and records findings in a living QA docs tree.
insta-master
Plans, scripts, distributes, and monetizes Instagram content using retention and SEO frameworks.
tanstack
Implements TanStack Query, Router, and Form patterns for React applications.
writing-agents-md
Write and trim lean AGENTS.md and CLAUDE.md agent instruction files.
architectural-analysis
Audits codebases for dead code, duplication, anti-patterns, type issues, and code smells.
bubbletea
Build terminal user interfaces in Go with Bubbletea, Lipgloss, and the Elm architecture.
vc-strategy
Designs startup fundraising strategy covering fundability, narrative, round sizing, and term-sheet negotiation.
spec-peer-review
Runs cross-LLM peer review of technical specs via compozy exec, producing validated Markdown findings artifacts.
yc-apply
Guides founders through a 10-phase Y Combinator application workflow with grilled drafts and gated submission.
vc-outreach
Executes VC fundraising outreach from investor targeting through term-sheet-to-wire close.
writing-tech-post
Drafts engineering blog posts using archetype contracts, evidence cadence, and pre-publish gates.
vc-pitch-deck
Builds and reviews investor pitch decks using slide sequences, proof standards, and teardown rubrics.
rust-best-practices
Applies Rust coding guidelines covering ownership, error handling, async patterns, testing, and linting.
qa-report
Plans persona-driven QA cycles as living repository documentation trees.
yt-master
Plans, scripts, packages, and optimizes YouTube videos using packaging-first methodology.
deep-review
Reviews branch diffs, worktrees, and GitHub PRs with parallel reviewer cohorts and verdict reports.
herdr-orchestration
Orchestrate Claude and Codex worker TUIs from a controller agent over the herdr socket CLI.
ui-craft
Design and audit user-facing UI surfaces against accessibility, state, and anti-slop standards.
deslop
Remove AI-generated code slop from branch diffs before commits and pull requests.
agent-exploration
Dispatches parallel scoped-write subagents to produce multi-area research analysis files.
testing-boss
Applies testing doctrine for authoring, reviewing, and gating software and LLM eval suites.
writing-skills
Author, refactor, and diagnose agent skills following the agentskills.io specification.