Help us improve
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
By queelius
Unified publication pipeline: R packages (CRAN/JOSS/JSS) and Python packages (PyPI)
npx claudepluginhub queelius/claude-anvil --plugin pub-pipelineAudit an R package against CRAN Repository Policy
Evaluate an R package against the JOSS reviewer checklist
Draft a JOSS paper (paper.md and paper.bib)
Publish a Python package to PyPI
Orchestrate the full R package publication pipeline
Reviews community and open-source practice against JOSS requirements: CONTRIBUTING.md, CODE_OF_CONDUCT.md, LICENSE, issue tracker accessibility, development history, and installation instructions. Lightweight audit focused on repository hygiene. Launched by the joss-reviewer orchestrator. <example> Context: Orchestrator needs community guidelines check. user: "Check if this repo has CONTRIBUTING and CODE_OF_CONDUCT" assistant: "I'll launch the community-auditor to check open-source practice requirements." </example> <example> Context: Package needs JOSS community requirements verified. user: "Does this repo meet JOSS community standards?" assistant: "I'll launch the community-auditor to evaluate community guidelines and repo hygiene." </example>
Searches for competing packages and related tools in the same domain as the package under review. Maps the "State of the Field" by finding CRAN/PyPI packages, GitHub projects, and published papers that address the same problem. Reports what each competitor does and what gap the reviewed package fills. Launched by the joss-writer or joss-reviewer orchestrator. <example> Context: Writer needs competing packages for State of the Field section. user: "Find competing R packages for survival analysis with hazard functions" assistant: "I'll launch the field-scout to map competing packages and identify gaps." </example> <example> Context: Reviewer needs to verify State of the Field completeness. user: "Are there packages missing from this paper's State of the Field?" assistant: "I'll launch the field-scout to find packages that should be mentioned." </example>
JOSS submission review orchestrator. Acts as a JOSS editor: reads the package and paper.md, spawns specialist auditors in parallel (software-auditor, community-auditor, field-scout), cross-verifies findings, and synthesizes a unified report against the JOSS reviewer checklist. Produces an actionable gap report with severity ratings and fix suggestions. <example> Context: User wants to check JOSS readiness before submitting. user: "Audit my package for JOSS submission" assistant: "I'll launch the joss-reviewer agent for a comprehensive JOSS readiness audit." </example> <example> Context: User has paper.md and wants it reviewed against JOSS standards. user: "Review my JOSS paper" assistant: "I'll launch the joss-reviewer agent to evaluate against the full JOSS checklist." </example>
JOSS paper drafting orchestrator. Reads the package source, README, vignettes, and existing manuscripts, spawns a field-scout to map competing packages, then drafts a complete paper.md and paper.bib following current JOSS format requirements (2025+). Produces publication-ready output with all required sections: Summary, Statement of Need, State of the Field, Software Design, Research Impact, AI Usage Disclosure, Acknowledgements, and References. <example> Context: User wants to draft a JOSS paper for their R package. user: "Draft a JOSS paper for flexhaz" assistant: "I'll launch the joss-writer agent to analyze the package and draft paper.md." </example> <example> Context: User has an existing manuscript to reshape into JOSS format. user: "Convert my paper to JOSS format" assistant: "I'll launch the joss-writer agent to reshape your manuscript into JOSS paper.md." </example>
Reviews software quality against JOSS functionality criteria: API design, documentation completeness, test coverage, CI/CD setup, installation, and error handling. Produces a structured findings report with severity ratings. Launched by the joss-reviewer orchestrator. <example> Context: Orchestrator needs software quality assessment. user: "Audit the software quality of this R package" assistant: "I'll launch the software-auditor to evaluate tests, docs, CI, and API design." </example> <example> Context: Package needs functionality verification before JOSS submission. user: "Does this package meet JOSS software requirements?" assistant: "I'll launch the software-auditor to check against JOSS functionality criteria." </example>
This skill should be used when the user asks to "audit for CRAN", "check CRAN readiness", "prepare for CRAN submission", "CRAN check", "is my package CRAN ready", "submit to CRAN", or mentions CRAN submission compliance. It audits an R package against CRAN Repository Policy and submission requirements, producing a structured gap report with actionable fixes.
This skill should be used when the user asks to "audit for JOSS", "check JOSS readiness", "JOSS reviewer checklist", "prepare for JOSS submission", "is my package JOSS ready", "JOSS requirements", or mentions JOSS submission compliance. It evaluates an R package and its paper.md against the complete JOSS reviewer checklist, producing a structured gap report.
This skill should be used when the user asks to "draft a JOSS paper", "write paper.md", "create JOSS paper", "write JOSS submission", "draft paper.md for JOSS", "convert paper to JOSS format", or mentions writing a JOSS software paper. It discovers existing manuscripts and package content, then drafts a complete paper.md and paper.bib following JOSS format requirements.
This skill should be used when the user asks to "publish my package", "publication pipeline", "get this published", "how do I publish this", or mentions publication without specifying an ecosystem. It detects the project type (R package or Python package) and routes to the appropriate ecosystem-specific publication skill.
This skill should be used when the user asks to "publish to PyPI", "upload to PyPI", "PyPI release", "publish my Python package", "package for PyPI", "submit to PyPI", "PyPI submission", or mentions PyPI publishing. It guides the complete workflow: auditing package metadata, running tests, building distributions, testing on TestPyPI, and publishing to PyPI.
Uses power tools
Uses Bash, Write, or Edit tools
Share bugs, ideas, or general feedback.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Research integrity plugin for Claude Code — paper auditing, citation verification, experiment analysis, and methodology-first skills for academic workflows.
Development automation skills for Python projects
Automates pubm setup for new open source projects — configures registries, CI, and changesets with a single skill invocation.
Multi-agent workflow framework for building, testing, and shipping statistical software packages
Research-team agents for Claude Code: supervisor, analysis-implementer, paper-writer, figure-descriptor, reviewer, literature-curator.
Manage your academic publication pipeline from Claude Code — list, search, create, move, analyse, export BibTeX, track reminders, and sync papers (GitHub + Overleaf) via the Kabbo MCP server.
Journal-aware Claude Code sessions. Surfaces relevant tasks, ideas, and plans from your jot journal on session start. Provides slash commands and deep journal analysis.
Cross-post blog content to multiple platforms — audit, publish, rewrite, and track content distribution via the crier CLI
Confidentiality toolkit — smart routing across cryptoid (Hugo encryption), pagevault (HTML/file encryption), and gpg (file encryption/signing)
Collection-aware repository intelligence for Claude Code. Query, analyze, and maintain your git repository collection.
Amazon KDP book publishing: manuscript audit, listing craft, and submission workflow
A forge for Claude Code plugins.
Claude Anvil is a plugin marketplace containing nine Claude Code plugins for academic research, creative worldbuilding, publication workflows, site management, repository intelligence, personal metadata, encryption, and more. Each plugin provides skills, commands, and agents that extend Claude Code with domain-specific capabilities.
# Install the marketplace
/plugin marketplace add queelius/claude-anvil
# Install an individual plugin
/plugin install papermill@queelius
| Plugin | Description | Version |
|---|---|---|
| papermill | Academic paper lifecycle: thesis, lit survey, experiment, review, venue, submission | 0.3.0 |
| worldsmith | Documentation-first fiction worldbuilding (the "Silmarillion approach") | 0.2.0 |
| pub-pipeline | Publication workflows: R/CRAN/JOSS, Python/PyPI | 0.5.0 |
| mf | Metafunctor site management: blog architecture, content workflows, crier | 1.0.0 |
| repoindex | Collection-aware repository intelligence — query, analyze, maintain git repos | 0.10.0 |
| deets | Personal metadata queries — identity, contact, academic, profiles | 1.0.0 |
| alex-confidential | Confidentiality toolkit — cryptoid, pagevault, gpg encryption | 0.1.0 |
| kdp | Amazon KDP book publishing: manuscript audit, listing craft, submission workflow | 0.1.0 |
| jot | Journal-aware sessions — surfaces tasks, ideas, and plans from your jot journal | 0.1.0 |
Each plugin follows Claude Code plugin conventions:
<plugin>/
├── .claude-plugin/plugin.json # Manifest: name, version, description, author
├── skills/<name>/SKILL.md # Interactive skills (the core logic)
├── commands/<name>.md # Slash commands (thin wrappers or rich docs)
├── agents/<name>.md # Autonomous subagents with system prompts
└── hooks/hooks.json # Event handlers (optional)
Skills are the heart of each plugin — they contain the domain knowledge and workflow logic. Commands are thin wrappers that trigger skills via /plugin:command syntax. Agents run autonomously for tasks like literature surveys or code review.
Not every plugin uses every component. Minimal plugins like deets have only a skill and a command. Larger plugins like worldsmith add hooks for ambient project detection and propagation discipline.
MIT