Brian Borghei
fe446a8a2e
Final layer of the depth investment. Deterministic scoring of any markdown artifact against per-skill rubrics. No LLM in the loop, stdlib only. FRAMEWORK (evals/) - engine.py — generic check evaluator with 13 check types: regex, regex_not, section_present, section_word_count, section_sentence_count, keyword_any, keyword_all, keyword_none, has_table, has_list, url_count, length_in_range, line_count_range. - run.py — runner with --all, --skill, --artifact, --format, --threshold, --output flags. - README.md — framework documentation + how to add a new rubric. RUBRICS (12 skills, 190+ criteria total) Each rubric anchors every criterion to a specific entry in the skill's references/red-flags.md or SKILL.md Success Criteria: - create-prd — 17 criteria scoring against 8-section PRD model - prfaq — 19 criteria scoring against Amazon Working Backwards - ai-feature-prd — 21 criteria scoring against eval/guardrails/model-fallback - brainstorm-okrs — 19 criteria scoring against SMART + Wodtke confidence - status-update — 19 criteria scoring against SBNR + R/Y/G + Asks - post-mortem — 19 criteria scoring against blameless + 5 Whys + owners - north-star-metric — 19 criteria scoring against NSM + input tree + counter - product-vision — 19 criteria scoring against Pichler / Raskin structure - pricing-prd — 20 criteria scoring against Westendorp + grandfathering - roadmap-comms — 19 criteria scoring against 3-variant audience fit - release-notes — 19 criteria scoring against value framing + categories - feedback-triage — 19 criteria scoring against Kano + RICE + ack templates SMOKE TEST (all 12 worked examples self-scored) - 6 skills @ 100/100 (ai-feature-prd, customer-feedback-triage, north-star-metric, post-mortem, product-vision, status-update-generator) - 6 skills @ 89-95/100 (brainstorm-okrs 95, pricing-prd 95, prfaq 94, roadmap-communication 94, release-notes 90, create-prd 89) - 12/12 pass at threshold 70 PM README + CHANGELOG updated. This commit completes the 3-tier PM depth investment: - Tier 1 (505cb8e): worked examples (54), data adapters (3), MCP tools (15) - Tier 2 (a8063e1): red-flag libraries (54), runnable pipelines (5) - Tier 3 (this): output eval harness (12 rubrics) + framework PM domain: 54 skills, 78 sub-files of worked examples, 54 red-flag libraries, 12 deterministic scorers, 5 chain pipelines, 3 live data adapters, 15 MCP tools. The skills moved from "documentation" to "production toolkit."
Claude Skills
AI skills for every team, not just engineering.
Try it now
No setup, any AI chat. Browse the skill library, click Try it now on any skill, and paste the prompt into Claude.ai, ChatGPT, or Gemini.
Save it to your AI. Click Save to my AI to install as a Claude Project or Custom GPT. Guided setup, no terminal.
For developers
npx @borghei/claude-skills add senior-fullstack
Auto-detects Claude Code, Cursor, Codex, Gemini CLI, Copilot, Windsurf, Cline, Aider, and Goose. Drops the skill into the right directory for your assistant. Node 18+, no Python required.
See docs/INSTALLATION.md for the Claude Code plugin, MCP server, manual install, and CLI tooling.
What you get
Ship a SOC 2 audit. Run a sprint retro. Draft a board update. Spin up a fintech compliance program. Real workflows you can finish today, not prompt templates that need a prompt engineer to drive them.
- The only OSS skills library covering regulated industries. SOC 2, ISO 13485, MDR, FDA, EU AI Act, NIS2, DORA, HIPAA, GDPR. 18 compliance frameworks with checklists, gap analyses, and audit-ready outputs.
- Scripts that actually run, not just prompt templates. 700+ Python tools you can pipe into CI, MCP servers, or your own scripts. Standard library only, no pip cliffs.
- Call one persona, get a stack of skills behind it.
startup-cto,growth-marketer,solo-founder,cs-cto-advisor,cs-fintech-advisor, and 70 more. 74 expert identities ready out of the box. - Works with your AI assistant. Claude Code, Cursor, Codex, Gemini CLI, Copilot, Windsurf, Cline, Aider, Goose, OpenCode, and ChatGPT / Claude.ai.
- Discovery → delivery → career for product teams. Project Management is the deepest domain (54 skills): PRDs, OKRs, status updates, story splitting, interview prep, with Jira / Linear / Notion / Confluence integrations.
Supporting evidence: 294 skills · 700+ Python tools · 74 agents · 18 compliance frameworks · 15 domains · 11 platforms.
Browse skills by domain
Project Management (42) ★ (most-visited domain. Discovery, delivery, career growth, Jira / Linear / Notion / Confluence.)
Engineering (76) · Marketing (38) · C-Level (26) · Compliance (21) · Business & Growth (16) · Legal (17) · Personal Productivity (10) · Product Team (8) · Vertical Advisors (7) · Data (5) · Sales (5) · HR (4) · Documents (4) · Finance (3) · All skills →
How it works
Pre-built expert identities
7 cross-domain personas that combine skills from multiple domains into a single expert:
startup-cto · growth-marketer · solo-founder · content-strategist · devops-engineer · finance-lead · product-manager
Activate with /persona startup-cto or just mention it in conversation. Add the 67 single-domain cs-* agents (e.g. cs-cto-advisor, cs-fintech-advisor, cs-scrum-master) for narrower expertise.
Deep-dive skill packs
Three multi-skill packs go deeper on a single topic: Playwright Pro (browser automation, 8 sub-skills), Self-Improving Agent (agents that evaluate their own outputs, 6 sub-skills), and AgentHub (multi-agent orchestration, 7 sub-skills). See standards/.
Slash commands
26 commands for common workflows: /tdd, /rice, /prd, /retro, /tech-debt, /security-scan, /sprint-plan, /changelog and more. Run /README in Claude Code for the full list, or see docs/commands/.
Ready-made kits by role
Pre-configured skill bundles for common roles:
SaaS Founder Kit · DevOps Kit · Compliance Kit · Growth Kit · Product Kit · Data Kit · Security Kit · Finance Kit · PM Starter Pack · PM Career Track
Install one bundle and the right set of skills lands in your project. See bundles.json.
Orchestration patterns
4 multi-agent patterns for complex workflows: sequential pipeline, parallel fan-out, supervisor delegation, and consensus voting. Defined in the Orchestration Protocol standard.
Works with your AI assistant
| Platform | Status |
|---|---|
| Claude Code | Native |
| OpenAI Codex | Native |
| Gemini CLI | Native |
| Cursor | Native |
| GitHub Copilot | Native |
| Windsurf | Native |
| Cline | Native |
| Goose | Native |
| Aider | Compatible |
| Jules | Compatible |
| RooCode | Compatible |
See docs/INSTALLATION.md for per-platform setup.
Documentation
- Full docs site: borghei.github.io/Claude-Skills
- Skills reference: docs/SKILLS.md
- Agents reference: docs/AGENTS.md
- Usage guide: docs/USAGE.md
- Installation: docs/INSTALLATION.md
- Contributing: CONTRIBUTING.md
- Standards: standards/
Contributing
Contributions welcome. See CONTRIBUTING.md for guidelines. Fork the repo, create a skill following the Skill Authoring Standard, include Python tools and YAML frontmatter, and submit a PR.
Contributors
| Contributor | GitHub |
|---|---|
| Alan Pope | @popey |
| Izzy | @weemax |
| Rohan (Tessl) | @rohan-tessl |
Disclaimer
This project is built with the assistance of AI tools (Claude, GPT, etc.). While every effort is made to ensure accuracy, AI-generated content (including skill definitions, reference guides, Python tools, frameworks, and documentation) may contain errors, inaccuracies, or outdated information. Always verify critical information independently before using it in production, compliance, legal, financial, or safety-critical contexts. The authors accept no liability for decisions made based on the content in this repository. Use at your own risk.
License
MIT + Commons Clause. Free for open-source, personal, education, and internal business use. Cannot be sold or repackaged as a paid product. See LICENSE for full terms.
294 skills · 700+ tools · 15 domains · 74 agents · 11 platforms.
borghei.me
