Wes Braga
Community@wesbragagt · Nashville, TN
It's not the destination but the homies we've made along the way. 🇧🇷 Brazilian, musician turned programmer.
Agent Skills by Wes Braga
Showing 36 vetted skills indexed across 1 GitHub repositories.
research
Researches topics using web search via exacli and presents cited findings.
commit
Stages only task-relevant files and creates conventional git commits.
markdown-review
Opens local Markdown files in a centered tmux Glow popup for review and editing.
prompter
Crafts structured prompts for delegating tasks to subagents using a seven-step framework.
notify
Sends desktop notifications with status, tmux context, and terminal-focus actions.
pr-feedback
Fetch, triage, and resolve open pull request review comments using the GitHub CLI.
video
Routes video creation requests to HyperFrames renders, demo GIFs, and media workflows.
deepwiki
Query GitHub repositories about code, architecture, and configuration using DeepWiki.
tasks
Manage spec.md and tasks.yaml task sets with status tracking and dependency queries.
learn
Generates a 20-hour structured learning plan of ten two-hour sessions for any topic.
coordinator
Orchestrate parallel git worktree agents via workmux CLI commands.
grillme
Interrogates a plan or design through sequential one-at-a-time questions until shared understanding is reached.
memory
Store and recall durable facts and decisions using the Mnemosyne CLI.
merge
Commits staged changes, rebases onto the base branch, and merges via workmux.
grilldebug
Diagnose bugs through one-question-at-a-time interrogation and systematic root cause analysis.
ddd
Design software architecture using Domain-Driven Design and Hexagonal Ports & Adapters patterns.
handoff
Delegates implementation tasks to a background agent in a new workmux worktree with a self-contained prompt.
exacli
Searches the web, code, and URLs through the Exa AI search API from the command line.
worktree
Dispatches tasks into new git worktrees using workmux with generated prompt files.
improve-codebase-architecture
Explores codebases to identify module-deepening refactors that improve testability and architecture.
rulereview
Reviews code changes against personal rule files using a prosecute-then-defend adversarial process.
code
Executes tasks from a tasks.yaml file by orchestrating code-writer and verifier agents.
to-spec
Creates implementation-ready specifications and self-contained task packets for agent-driven development.
excalidraw
Create hand-drawn Excalidraw diagrams from JSON using the excalidraw-cli command-line tool.
Frequently Asked Questions About Wes Braga
FAQPage SchemaWhat tasks can I accomplish with Wes Braga's skills?▼
You can manage git commits, rebases, merges, and pull requests; orchestrate parallel agents in worktrees; apply DDD and hexagonal architecture; run TDD cycles; review PR feedback; research topics via web search; create Excalidraw diagrams; and manage tmux sessions non-interactively.
Who are these skills designed for?▼
Software engineers and developers working in terminal-centric environments who use git heavily, delegate work to coding agents, practice test-driven development, and want structured architecture guidance such as Domain-Driven Design and ports-and-adapters patterns.
How do the git and PR skills work in practice?▼
The commit skill stages only task-relevant files, never using git add -A. The pr skill commits pending changes, creates or updates a GitHub pull request, then monitors CI checks and reports failures. The rebase skill walks through conflict resolution step by step.
What prerequisites or dependencies do these skills require?▼
Several skills depend on specific utilities: workmux for worktree management, tmux for terminal sessions, Glow for Markdown review, excalidraw-cli for diagrams, Mnemosyne CLI for memory, uvx with ask-deepwiki for repository Q&A, and the Exa search CLI for web research.
How do the orchestration skills coordinate multiple agents?▼
The coordinator skill spawns, monitors, and merges worktree agents; handoff delegates tasks to a fresh background agent with a self-contained prompt; the code skill executes tasks.yaml by orchestrating code-writer agents; and adreview runs adversarial review dialogues between sibling agents.