claude-marketplace

Manage a local task/backlog list (tasks.json + a generated TASKS.md) for the current repo and reconcile it with GitHub or Gitea issues, via the bundled task-sync Python tool. Supports direct commands (list/add/edit/done/remove/ status/init) and a plan-decide-apply sync that surfaces creates, pushes, pulls, conflicts, and confidentiality findings for review before anything leaves the machine, with a public-repo push warning and a --dry-run mode. Suggest when — the user wants to see or manage this repo's tasks or backlog, sync tasks with GitHub/Gitea issues, mentions a local task tracker or TASKS.md, or asks to push/pull/reconcile tasks against issues; keywords — task list, backlog, sync issues, task-sync, todo list, reconcile issues.

Initialize mandatory experiment logging using scientific notebook, ADR, and postmortem patterns. For projects involving system changes, benchmarks, debugging, or exploratory work. Creates LAB_NOTEBOOK.md with living decision/action tracking and injects iron-clad CLAUDE.md rules that make logging a PRECONDITION for every action. Suggest (do not auto-run) when — infrastructure/experimental/expensive-failure projects, multi-session work, or keywords like optimize/benchmark/debug/configure.

Generate BPMN 2.0 compliant XML files from natural language process descriptions OR from structured markdown business process documents. Use this skill when a user wants to create a BPMN workflow, convert a business process to BPMN XML, model a workflow diagram, or generate process definitions. Triggers on requests like "create a BPMN", "generate workflow XML", "model this process", "convert to BPMN 2.0", "create process diagram", "build workflow", or "convert this markdown to BPMN". Do NOT use for converting an existing BPMN XML file to Draw.io — use bpmn-to-drawio for that.

Analyze technical documents for CS/ML concepts a smart non-CS reader wouldn't understand, recommend explanation mechanisms (glossary, inline, footnote, sidebar, appendix), present analysis for approval, then implement annotations in the Word document using the project directory for contextual accuracy. Use when making technical documents accessible to non-CS audiences. Do NOT use for writing a new overview document from a codebase — use explain-project for that.

Comprehensive 9-agent architecture review — spawns parallel domain specialists (architecture, code, data, integration, performance, QA, security, platform, risk) and produces structured findings with executive report and go/no-go recommendation

Set up a persistent, LLM-maintained wiki inside any project. Creates a wiki/ directory with sources, pages, schema, and navigation files, seeds initial pages from project discovery, and injects CLAUDE.md rules that make Claude automatically maintain the wiki during normal work sessions. Suggest (do not auto-run) when — project accumulating complexity, context loss between sessions, multiple contributors, "I keep forgetting", significant domain knowledge, or LAB_NOTEBOOK.md with durable insights.

Thoroughly evaluate contact-center-lab pipeline output quality against input, checking sanitization correctness, atom/entity/triple quality, graph structure, and procedure integrity across all stages. Suggest when — pipeline run complete and quality evaluation needed, "did this produce good output" questions, before full corpus run, or after pipeline code changes.

Generate a comprehensive, annotated technical overview document for any project/repo, written for a smart non-CS reader. Analyzes the codebase, writes a deep-dive document following a proven structure template, and produces a styled Word document with sidebars, glossary, inline annotations, and optional generated images. Use when a project needs an explanatory document that makes the system understandable to non-technical stakeholders. Do NOT use for annotating an EXISTING document — use accessibility-annotator for that. Do NOT use for plain markdown-to-Word formatting with no annotations — use /convert-markdown for that.

One-shot, read-only health snapshot across the personal fleet (DGX Spark, Jetson Orin Nano, homeserver, bond, obvm). Checks uptime, load, disk, memory, and each machine's key inference/service endpoints over SSH and curl, then renders a single status table with a pass/fail verdict. Use for a fleet status check, a morning brief before starting work, answering "is the spark available" or similar per-machine availability questions, or any general machine health check across the 5 hosts. Read-only — never restarts, stops, or reconfigures anything on any host.

Use when checking on Jetson Orin Nano inference performance landscape — scans JetPack updates, llama.cpp releases, small model landscape, NVIDIA Jetson forum, and live device health for actionable changes. Run periodically from the jetson project directory.

Audit a dataset for proprietary information leaks before sharing with public/cloud services. Produces a LEAK_RISK.md report with findings and remediation recommendations.

SSH into the DGX Spark and audit all running containers against known best practices and community optimizations. Reports gaps, misconfigurations, and optimization opportunities. Complements spark-recon (external landscape) with internal config validation.

Use when checking on DGX Spark inference performance landscape — scans Arena leaderboard, vLLM releases, spark-vllm-docker builds, Qwen models, and NVIDIA forum for actionable changes. Run periodically from the spark project directory.

Use when the user has a spec document, design system reference, component library doc, wireframe description, or similar specification and wants a visual HTML/CSS dummy prototype built from it. Triggers on "build a prototype", "create a mockup from this spec", "prototype this design", "make a visual demo". Also use when converting technical documentation into stakeholder-ready visual demos. Do NOT use for production frontend implementation — this produces visual HTML/CSS dummies only, not shippable code.

Structured implementation planning for bug lists, feature requests, or change sets. Enforces deep investigation, interaction mapping, and integrated solution design before any code changes. Use when given a list of issues/features/changes that need a cohesive implementation plan.

Wiki operations: ingest source documents into wiki pages, lint for health issues, query the wiki for answers, propagate resolved facts, and report status. Supports both /create-wiki layouts (wiki/ + schema.yaml) and OKF-bundle layouts (kb/ + AGENTS.md contract).

Assemble final PowerPoint (.pptx) from presentation markdown and generated images. Use when you want to create the final PowerPoint, build the deck after image generation, or skip to building from presentation.md.

Draft full slide content (titles, bullets, speaker notes, graphics descriptions) from an outline. Use when you have an outline and want to generate full slide content, draft the slides, or move from structure to full presentation content.

Run the complete 7-step slide generation pipeline from topic to PowerPoint. Use when you want to create a full presentation from scratch, generate a complete deck end-to-end, or see the entire pipeline in action.

Generate slide visuals using Gemini Pro from validated graphics descriptions. Use when you want to generate slide images, create visuals after validation, or see what the slides will look like.

Run quality analysis and automated improvement on drafted slide content. Use when you have drafted slides and want to improve quality, optimize readability, or get quality scoring before finalizing.

Generate a structured presentation outline from research findings. Use when you have research.json and want to create an outline, structure findings into slides, or plan slide structure before drafting.

Conduct autonomous web research on a topic using Claude Agent SDK, producing structured research.json. Use when researching a topic for slides, starting the pipeline from scratch, or you have a topic but no research.json yet.

Validate that image descriptions are concrete enough for AI image generation. Use when you want to check if image descriptions are ready for generation, validate graphics before image generation, or avoid wasted API calls.

Send a capture to Open Brain. Takes any instruction as an argument — summarize a session, log a decision, capture an idea — processes it, and POSTs to the captures API. Use when the user says "/brain-entry <instruction>". Suggest (do not auto-run) when — user says "log/capture/remember/brain entry", end of session, decision to record, or explicit invocation.

Archive, retire, or sunset a project repo — writes a status header into README.md, tags and commits the change, pushes and optionally archives the remote (GitHub only), moves the directory into ~/dev/archive/, and logs one line to ~/dev/PORTFOLIO.md. Suggest when the user says archive project, retire project, sunset repo, mothball this repo, shelve this project, or asks to wind down/decommission a repo that is no longer active.

SSH into the Jetson Orin Nano and audit the running inference config against known best practices and community optimizations. Reports gaps, misconfigurations, and optimization opportunities. Complements jetson-recon (external landscape) with internal config validation.

Scaffold a brand-new project directory end-to-end — git init, remote repository (GitHub by default, Gitea with --gitea), CLAUDE.md, .gitignore, placeholder .env, lab notebook, kill-criteria BRIEF.md, and the initial commit/push. Suggest (do not auto-run) when the user says new project, scaffold project, start a project, or asks to set up a fresh repo under ~/dev.

Orchestrate parallel deep research across multiple LLM providers using native context:fork subagents and synthesize results. Suggest when — in-depth topic research, multi-provider perspective comparison, well-sourced analysis needed, "deep research"/"research report" keywords, or thorough technical/strategic/emerging topic investigation.

Comprehensive security analysis with tech stack detection, vulnerability scanning, and remediation planning. Suggest when — security/vulnerabilities/CVEs/audit mentioned, new projects scaffolded, before releases/deployments/production, auth/input-handling code review, dependency updates, or new repos cloned.

Transform text or documents into AI-generated infographic pages that explain concepts visually using Gemini Pro 3 for generation and Claude Vision for quality evaluation. Suggest when — user has document/report/concept to visualize, after report generation, infographic/visual explanation keywords, improving visual appeal, or whitepaper/guide transformation.

Prepare a project for a context clear (/clear) or compaction with zero loss of state. Flushes the current session's work into all persistent documents — LAB_NOTEBOOK.md living sections and in-flight entry, memory files, CLAUDE.md rules, CHANGELOG — then emits a single copy-paste resume prompt to run in the fresh session so a zero-context Claude picks up exactly where you left off. Suggest (do not auto-run) when the user says they want to clear context, compact, wrap up the session, or hand off.

{{DESCRIPTION}}

Convert BPMN 2.0 XML files into Draw.io native format (.drawio) using the bpmn2drawio Python tool. Renders properly in Draw.io Desktop or web applications. Use this skill when a user wants to visualize a BPMN process in Draw.io, convert BPMN to editable diagrams, or create Draw.io files from process definitions. Triggers on: "convert BPMN to Draw.io", "create drawio from BPMN", "visualize BPMN in Draw.io". Do NOT use for creating new BPMN from a process description or markdown doc — use bpmn-generator for that.

Before starting complex multi-step implementation tasks, assess scope and route to the right planning approach — native plan mode for simple changes, /plan-improvements for codebase refactoring, or /create-plan for requirements-driven work. Suggest when — implementation task with >3 files, architectural decisions, ambiguous scope, or multiple valid approaches exist — and no active IMPLEMENTATION_PLAN.md.

Evaluate an existing codebase to produce a detailed report on project purpose, health, status, and recommended next steps. Suggest when — new session project questions, encountering repo first time, resuming after break, health/quality/documentation inquiries, architectural decisions, or "overview"/"assess" requests.

Validate plugins, clean repository, and ship plugin releases in one automated workflow. Suggest when — plugin changes complete and ready to ship, after validate-plugin succeeds, before releases to ensure quality, "validate and ship" keywords, or batch improvements done.

Create branch, commit, push, open PR, auto-review, fix issues, and merge — the full ship workflow. Suggest when the user signals completion (done, ready to ship, push this, let's ship it, or similar), after a work item from an implementation plan finishes, after all tests pass and changes are ready, when the user asks to create a PR or push changes, or after a code review cycle completes and changes are approved.

Synthesize employee feedback from Notion Voice Captures into a professional .docx assessment document. Suggest when — user mentions feedback analysis, performance review, assessment generation, Notion captures, or review preparation.

Load secrets from Bitwarden Secrets Manager into environment using bws CLI. Fully stateless — no vault unlock or session tokens required. Suggest (do not auto-run) when — session start with API keys needed, before skills requiring keys (research-topic, visual-explainer, summarize-feedback), Bitwarden/secrets/API key mentions, or auth errors.