maragu
Official@maragudk · Aarhus, Denmark
Provides specialized Go-based web development, architectural documentation, and iterative codebase maintenance systems for enterprise engineering teams.
Agent Skills by maragu
Showing 35 vetted skills indexed across 3 GitHub repositories.
gardeners
Coordinate parallel autonomous gardeners to fix codebase issues via a shared task list.
autoresearch
Automate iterative experiments to optimize a measurable metric.
code-reviewers
Coordinate two independent reviewers to challenge findings and produce a consolidated diff report.
design-doc
Draft structured design documents with goals, constraints, decisions, and acceptance criteria.
improve-skill
Collect conversation signals and draft PR-ready edits to skills/*/SKILL.md.
garden
Scan a codebase for maintenance issues and fix one per run.
gomponents
Construct type-safe HTML components with a Node-like API in Go.
spec
Create and update a living product spec at docs/spec.md.
observable-plot
Generate charts from JavaScript data using Observable Plot's declarative API.
datastar
Build server-driven reactive frontends for Go web apps with SSE updates.
atproto
Describes how to build on the AT Protocol with Lexicons, XRPC, and firehose.
bluesky
Publish Bluesky posts with text, images, videos, replies, and quotes.
marimo
Create and document marimo Skill units storing notebooks as pure Python files.
blog-post-interview
Conduct a one-question-per-message interview to define a blog post's angle, audience, and evidence.
diary
Automate structured implementation diary entries with frontmatter and step-wise guidance.
modal
Execute Python code on the Modal serverless compute platform.
dad-joke
Deliver a single dad joke with a setup, punchline, and explanation.
address-code-review
Triage code-review comments sequentially before implementation and record decisions.
security-review
Trace imports, callers, and data flow to report the most significant security issue.
unsloth
Fine-tune open LLMs on a single GPU with memory-efficient workflows.
failure-taxonomy
Builds a structured taxonomy of failure modes from open-coded trace annotations using axial coding.
llm-as-a-judge
Build and validate LLM-as-Judge evaluators for automated quality assessment.
prompt-engineering
Crafts, reviews, and refines prompts for LLM pipelines, including system and LLM-as-Judge prompts.
trace-annotation-tool
Generate a FastHTML web app for annotating LLM traces with pass/fail/defer status.
Frequently Asked Questions About maragu
FAQPage SchemaWhat specific tasks can be performed using these development systems?▼
These systems enable structured architectural decision logging, type-safe HTML component construction in Go, reactive server-driven frontend updates, and systematic codebase maintenance through autonomous issue tracking and resolution.
Which engineering personas benefit most from these capabilities?▼
Full-stack engineers, Go developers, and technical architects focused on maintaining high-integrity codebases, enforcing strict documentation standards, and building reactive web interfaces without heavy client-side dependencies.
What are the primary prerequisites for implementing these development patterns?▼
Implementation requires a Go development environment, familiarity with SQLite for local data persistence, and an understanding of server-side reactive patterns. Specific components like Datastar or Gomponents require standard Go module integration within your existing project structure.