Maximilian Wolf
Community@MaxWolf-01 · Vienna
Maximilian Wolf maintains a 45-skill Claude Code skill library covering session orchestration, code review, ticket dispatch, testing, and ML project conventions.
Agent Skills by Maximilian Wolf
Showing 45 vetted skills indexed across 1 GitHub repositories.
MaxWolf-01pocock-sync
Reviews upstream mattpocock/skills commits and adopts fitting changes into adapted mx skills.
MaxWolf-01research
Investigate questions against primary sources and write numbered research artefact files.
MaxWolf-01recap
Generates a structured status report of findings, decisions, open questions, and next steps.
MaxWolf-01fork
Spawns a subagent that inherits the full conversation transcript for context-heavy delegated tasks.
MaxWolf-01restore-sessions
Scan Claude Code session files and classify unfinished work for resuming.
MaxWolf-01writing-for-humans
Reviews and rewrites artifact prose against a catalogue of AI writing tells.
MaxWolf-01dependabot-triage
Triage open Dependabot alerts by tracing dependency usage and dismissing non-exploitable noise.
MaxWolf-01permissions-review
Review session transcripts and update Claude Code's Bash command permission allowlist.
MaxWolf-01writing-for-agents
Guides writing and structuring documents that AI agents consume, including skills, AGENTS.md, and specs.
MaxWolf-01orient
Routes agent work to the correct mx skill or workflow based on the current situation.
MaxWolf-01domain-modelling
Build and maintain project domain glossaries and architecture decision records in Markdown.
MaxWolf-01changelog
Updates CHANGELOG.md with notable changes from git commits since the last release tag.
MaxWolf-01wait-what
Rephrases the previous agent message in ASD-STE100 Simplified Technical English using project terminology.
MaxWolf-01handoff
Writes a handoff document compacting the current conversation for a fresh agent session.
MaxWolf-01gh-stack
Manages stacked pull requests and splits multi-part work into reviewable branches with gh-stack.
MaxWolf-01dispatch
Orchestrates worker agents across git worktrees to build and integrate ticketed feature work.
MaxWolf-01diagnosing-bugs
Diagnose hard bugs and performance regressions through a six-phase feedback-loop workflow.
MaxWolf-01transcript
Saves the current session transcript to a dated file for later continuation.
MaxWolf-01code-review
Reviews git diffs against a fixed point using parallel correctness, standards, spec, and test reviewers.
MaxWolf-01show
Create diagrams, comparisons, demos, and explainer pages as visual explanation artifacts.
MaxWolf-01to-tickets
Breaks a grilled specification into dependency-ordered tracer-bullet ticket files with blocking edges.
MaxWolf-01codex
Runs OpenAI Codex CLI non-interactively to obtain second-opinion code reviews and design feedback.
MaxWolf-01wizard
Generate interactive bash wizards that walk humans through manual setup and migration procedures.
MaxWolf-01improve-codebase-architecture
Scan a codebase for shallow modules and present deepening refactors as a visual HTML report.
Frequently Asked Questions About Maximilian Wolf
FAQPage SchemaWhat tasks can I accomplish with Maximilian Wolf's skill collection?▼
The 45 skills cover session handoff and forking, stacked PR management, parallel code review, ticket dispatch to worker agents, bug diagnosis loops, changelog generation, Dependabot triage, bloat audits, mermaid diagrams, uv script conversion, and ML experiment conventions.
Who is the target audience for these skills?▼
Software engineers running Claude Code who manage multi-session projects, stacked pull requests, and ticketed feature work. It also serves ML practitioners needing experiment tracking conventions and maintainers writing docs, ADRs, and READMEs consumed by both humans and agents.
How do I install and run these skills?▼
Place the skill folders into a Claude Code skills directory; each folder's frontmatter declares its name, description, and invocation rules. Skills like orient route you to the right one, while user-invocable entries such as session-name and writing-for-humans can be triggered directly.
Are these skills free and open source?▼
The skills are published publicly on Maximilian Wolf's GitHub account (MaxWolf-01) across 36 public repositories. No license or pricing is stated in the manifest, so check the individual repository for licensing terms before redistribution or commercial use.
What prerequisites and dependencies do these skills require?▼
Skills run inside Claude Code sessions. Specific entries declare allowed commands: dependabot-triage needs gh, npm, and uv; transcript needs ccx; pr-tldr needs Bash. The tmux skill requires tmux for long-running builds, tests, and training runs, and uv-script targets PEP 723 inline metadata.