Profile My skills Saved Collections Edit profile Submit a skill

← all publishers

datashaman

@datashaman source repo

12 published skills

  1. ▌
    Harness · datashaman bundle
    Control surface for a harness-engineering Claude Code setup at user or project scope. Handles install, uninstall, update, doctor, adopt, snapshot, status, audit, and memoize. Installs CLAUDE.md, guardrail hooks, /verify, /plan, /critique, memory seeds, and settings patches; can also retrofit an existing project with scripts/harness-check.sh, report install state, snapshot ~/.claude/, and run deterministic memory hygiene. All sub-actions are idempotent. Use when asked to "set up my Claude Code", "install harness", "uninstall harness", "update harness", "diagnose my setup", "adopt harness", "retrofit", "snapshot my setup", "audit my setup", "harden my Claude", "memoize", "consolidate memory", or "prune memory".
    0
    installs
  2. ▌
    Onepager · datashaman bundle
    Generate a self-contained single-file HTML demo or presentation using Alpine.js + Tailwind via CDN. Emailable, double-clickable, no build step or server. Use when the user asks for a demo, prototype, slide deck, walkthrough, pitch, or interactive mockup. Triggers on phrases like "make a demo", "build a one-pager", "prototype this", "presentation/slides", "single-file HTML", or "no build step".
    0
    installs
  3. ▌
    Audit Docs · datashaman bundle
    Documentation health audit for projects and docs sites. Accepts a URL or local directory. Uses a live browser for URLs — handles rendered docs sites and SPAs. Scores doc hygiene (README, CONTRIBUTING, CHANGELOG, broken links, staleness), Diataxis classification (tutorials / how-to / reference / explanation), API/docstring coverage, documentation-site structure (navigation, search, dead links), and agent-readiness (CLAUDE.md, AGENTS.md, ADRs, copy-paste-safe examples, unambiguous terminology). Returns a 0–100 score with actionable findings. Use when the user says "audit the docs", "documentation review", "docs health check", or "is my documentation agent-friendly".
    0
    installs
  4. ▌
    Audit Jira · datashaman bundle
    Jira project health audit via the Atlassian MCP. Identifies velocity trends, backlog rot, bug clusters, assignee concentration risk, and firefighting patterns. Requires the Atlassian MCP to be connected. Use when asked to "audit jira", "project health", "backlog analysis", or "team velocity".
    0
    installs
  5. ▌
    Audit Design · datashaman bundle
    Design and WCAG accessibility audit for web UIs. Accepts a URL or local directory. Uses a live browser for URLs — handles SPAs and dynamically-rendered content. Scores contrast, colour signalling, alt text, semantic structure, keyboard access, and other WCAG criteria, plus design hygiene (fonts, spacing, AI-slop patterns, component health). Returns a 0–100 score with actionable findings. Use when the user says "audit the design", "WCAG audit", "accessibility check", or "design review".
    0
    installs
  6. ▌
    Audit Trello · datashaman bundle
    Trello board health audit. Identifies throughput trends, backlog rot, blocker clusters, member concentration risk, and overdue patterns. Use when asked to "audit trello", "board health", "backlog analysis", or "what's overdue".
    0
    installs
  7. ▌
    Plan Website · datashaman
    Creates a Work Breakdown Structure for planning and executing a website build from scratch or as a migration/overhaul. Organizes platform and stack requirements, repository setup, planning artifacts, implementation slices, QA gates, and launch work into deliverables, work packages, and tasks, with phase checklists as supporting artifacts. Use when the user mentions building, launching, or overhauling a website: redesigns, CMS migrations, new site builds, or multi-team web projects. Triggers on phrases like "website build", "site launch", "web project checklist", "website redesign", "launching a site", "new website", or "website planning". Also use when the user wants to know what documents or deliverables a web project needs, or is coordinating design, dev, content, and SEO teams on a site.
    0
    installs
  8. ▌
    Audit Context · datashaman bundle
    Audit your Claude Code setup for token waste and context bloat. Checks MCP servers, CLAUDE.md files, skills, agents, commands, plugins, hooks, and permissions. Mines session transcripts for cache hit rate, autocompact frequency, turn costs, tool errors, and unused assets. Flags broken MCP servers that load schemas but never connect. Returns a health score with specific fixes. Use when asked to "audit my context", "check my settings", "token optimization", or "why is Claude slow".
    0
    installs
  9. ▌
    Audit Codebase · datashaman bundle
    Git-based codebase health audit. Identifies churn hotspots, bus factor risks, bug clusters, and firefighting patterns from git history. Use when asked to "audit the codebase", "codebase health", or "who owns this code".
    0
    installs
  10. ▌
    Workflow Advisor · datashaman bundle
    Interview a team about its development process, generate a versioned `.workflow/` configuration, and advise on GitHub events, lifecycle gates, labels, roles, and process metrics. Use when the user wants to set up team workflow automation, adopt spec-driven development, configure GitHub workflow files from an interview, design event-driven CI/CD around roles and artifacts, reconcile workflow state, or measure process changes. Triggers on "workflow advisor", "set up team process", "configure SDD", "process automation", "team workflow", "GitHub workflow generation", "spec-driven development setup", or bootstrapping repo process.
    0
    installs
  11. ▌
    Audit AI Strategy · datashaman bundle
    Audit a codebase's AI strategy through John Cutler's four-bucket lens: bad ideas amplified, good ideas supercharged, genuinely new possibilities, and the meta-skill of reading context. Identifies where AI is bolted onto broken patterns, where it amplifies what already works, and where the codebase could embrace workflows that only exist because AI is in the loop. Use when asked to "audit AI strategy", "evaluate our AI playbook", "find AI opportunities", or "where can we think outside the box with AI".
    0
    installs
  12. ▌
    Website Spec Pipeline · datashaman bundle
    Build a multi-stage website specification pipeline that moves from discovery through IA, page planning, UX structure, design system, responsive rules, wireframes, and review using JSON contracts between stages. Use when the user wants to design a website through agent handoffs, create auditable page specifications, generate downstream UI from structured artifacts, or avoid burying site architecture inside prompts.
    0
    installs