gerdsenai-marketplace

GerdsenAI MD-to-PDF Suite - Claude Code Plugin

A Claude Code plugin for autonomous sprint execution, adversarial analysis, deep research intelligence, and professional PDF generation. Powered by the GerdsenAI Document Builder.

What It Does

• Autonomous Sprint Execution (Autocoder) — plan and execute entire development sprints using the Socratic Method. Writes thousands of lines of code, auto-commits at planned points, manages context via vector DB for resilience against compaction. /gerdsenai:sprint-execute
• Adversarial Red-Team Analysis — 11-domain adversarial engine (code, security, deps, architecture, testing, DevOps, database, AI/ML, accessibility, documents, strategic) with Socratic reasoning chains and rabbit hole investigation. /gerdsenai:red-team
• Dual Vector DB Management — run ChromaDB (local) and Pinecone (cloud) simultaneously with configurable re-ranking, embedding models, and automated hooks (auto-upsert on commit, health checks on session start). /gerdsenai:vector-db
• Deep Research Intelligence Reports — multi-source research with parallel sub-agents, Mermaid visualizations, academic citations, adversarial quality review, and source freshness monitoring. /gerdsenai:research-report
• Professional PDF Generation — styled code blocks, cover pages, table of contents, headers/footers, Mermaid diagrams, logos, and flexible output locations. /gerdsenai:build
• Guided Setup — install, configure, or update the Document Builder with preferences for output, logos, page size, vector DB backends, and hooks. /gerdsenai:setup

Prerequisites

• Claude Code CLI installed
• Python 3.9+
• Windows: Git for Windows (provides Git Bash, which all scripts run under)
• Optional: Pinecone API key (for cloud vector DB), Ollama (for local AI inference)

Install

From within Claude Code (recommended)

1. Open Claude Code in any project
2. Run /plugin to open the plugin manager
3. Go to the Marketplaces tab
4. Add the marketplace: GerdsenAI/GerdsenAI-Markdown-To-PDF-Suite-Claude-Plugin
5. Go to the Discover tab
6. Select gerdsenai and install (User scope recommended)
7. Run /reload-plugins to activate

Local development

claude --plugin-dir /path/to/this/repo

After installing, run /gerdsenai:setup in any project to configure the Document Builder, vector DB backends, and preferences.

Quick Start

1. Set up the Document Builder (guided setup with preferences):

   /gerdsenai:setup
   

2. Build a PDF:

   /gerdsenai:build my-document.md
   

3. Research a topic and generate an intelligence report:

   /gerdsenai:research-report AI chip market landscape
   

Commands

Command	Description
/gerdsenai:setup	Install, configure, or update the Document Builder
/gerdsenai:build <target>	Build PDFs — single file or recursive directory (auto-detects)
/gerdsenai:research-report [topic|file]	Deep research + intelligence reports; source monitoring when given a file path
/gerdsenai:red-team <target>	Adversarial analysis: 11 domains (code, security, deps, architecture, testing, DevOps, DB, AI/ML, a11y, docs, strategic) with Socratic reasoning
/gerdsenai:vector-db [operation]	Vector DB management: report, store, query, configure — dual-backend (ChromaDB + Pinecone) with hooks
/gerdsenai:sprint-execute [plan|description|resume]	Autonomous sprint executor: Socratic planning, autocoding, auto-commit

Build Options

/gerdsenai:build report.md                        # Single file
/gerdsenai:build report.md --output-dir ~/Reports  # Custom output directory
/gerdsenai:build report.md --output-name Q4-Review  # Custom filename
/gerdsenai:build ./docs                            # Recursive directory build
/gerdsenai:build ./docs --recursive                # Explicit recursive flag

Recursive mode auto-excludes node_modules/, .git/, venv/, __pycache__/, .claude/, and common non-document files (README.md, CLAUDE.md, etc.).

Skill: PDF Document Authoring

The pdf-document-authoring skill activates when you're writing markdown intended for PDF output. It guides you on:

• YAML front matter fields (title, subtitle, author, date, version, confidential, watermark)
• Document structure and heading hierarchy
• Assessment/maturity sections using proper headings (not inline bold)
• Code block language identifiers for styled output (diff, tree, shell, python, yaml, etc.)
• Mermaid diagram syntax and best practices
• Table and image formatting
• Output location and filename options
• Quality checklist before building

Agents

Document Builder

Handles the full document creation workflow autonomously: