Lucas Faria
Community@lucasheriques · Belo Horizonte, Brazil
Lucas Faria maintains 23 Claude Code skills for adversarial code review, parallel subagent orchestration, decision logging, and TypeScript engineering rigor.
Agent Skills by Lucas Faria
Showing 23 vetted skills indexed across 1 GitHub repositories.
lucasheriquessync-upstream
Merge upstream cursor/plugins pstack changes into the Claude Code port via three-way git merge.
lucasheriquesfigure-it-out
Designs an auditable multi-phase playbook for engineering tasks matching no existing playbook.
lucasheriquesbro
Restates the previous AI message in plain language without jargon.
lucasheriquesinterrogate
Spawns multiple LLM reviewers to adversarially review code diffs and synthesizes a categorized verdict.
lucasheriquesswarm
Fan out parallel background workers and aggregate their results into one report.
lucasheriquestechnical-writing
Applies Diátaxis, Google style, STE, and Global English rules to technical documentation.
lucasheriqueswhy
Investigate design rationale behind code by querying git history and MCP evidence sources in parallel.
lucasheriqueshow
Generates architectural explanations of codebases by tracing subsystems with parallel explorer agents.
lucasheriquesblast-radius
Identifies what a code change could break beyond the diff and proves safety by running real code.
lucasheriquesreflect
Spawns three parallel review subagents over a session transcript to route learnings into skill edits.
lucasheriquesshow-me-your-work
Maintains an append-only TSV decision log with evidence for long-running or unattended work.
lucasheriquespoteto-mode
Routes engineering tasks to playbooks with principles, subagent delegation, and verified delivery.
lucasheriquesrecall
Reconstructs recent working context from chat transcripts and shared records into a status brief.
lucasheriquesmaintain-verification-skill
Audits a project's verification skill and feature map against source code and live app behavior.
lucasheriquestypescript-best-practices
Applies TypeScript type-safety rules and patterns when reading or editing .ts and .tsx files.
lucasheriquesarchitect
Sketch types, signatures, and module structure before implementing code.
lucasheriquesautomate-me
Drafts a personal -mode skill from mined conversation transcripts and user preferences.
lucasheriquesteach
Explains code changes and subsystems through progressive diagrams and plain-language walkthroughs.
lucasheriquesarena
Spawns parallel candidate solutions, judges them against a rubric, and synthesizes the strongest result.
lucasheriquescreate-verification-skill
Generates a project-local verification skill that drives an app like a user.
lucasheriquesno-comments
Removes comments added by a code change using an adversarial review subagent.
lucasheriquesunslop
Rewrites text to remove AI writing patterns and jargon.
lucasheriquestdd
Writes a failing regression test before fixing a bug, then verifies it passes after the fix.
Frequently Asked Questions About Lucas Faria
FAQPage SchemaWhat tasks can I accomplish with Lucas Faria's Claude Code skills?▼
You can run adversarial multi-model PR reviews, trace blast radius of risky diffs, orchestrate parallel subagent swarms, design module architecture before coding, enforce TypeScript best practices, write Diátaxis-structured docs, and maintain auditable TSV decision logs for long-running or unattended engineering work.
Who are these skills designed for?▼
Software engineers using Claude Code who want rigorous, verifiable workflows: reviewers needing pre-PR gates, developers resuming context after stepping away, architects sketching types before implementation, and teams requiring cited design rationale, postmortems, and proven verification of UI, CLI, or service behavior.
How do I invoke these skills in practice?▼
Most skills trigger via slash commands like /swarm, /arena, /architect, or /figure-it-out, or natural phrases such as 'blast radius of X' and 'catch me up'. The typescript-best-practices skill auto-applies to any .ts or .tsx file via path matching; unslop always applies to writing.
What prerequisites or dependencies do these skills require?▼
Skills run inside Claude Code and several depend on each other: teach composes how and why, figure-it-out logs via show-me-your-work, and automate-me uses skill-creator plus unslop. The why skill discovers available MCPs covering source control, issue trackers, chat, observability, and analytics.
How do these skills handle verification and code quality?▼
create-verification-skill generates a project-local control skill that drives your app like a user across any language or platform, while maintain-verification-skill audits it with parallel source readers and one live session. tdd applies only on explicit request, and blast-radius proves safety by running real code.