alignment-values-north-star
About
This skill helps teams establish shared values, decision-making frameworks, and strategic direction. It provides guidance for creating alignment artifacts like mission statements, cultural principles, and North Star metrics. Use it when starting new teams, scaling organizations, resolving misalignment, or defining product vision.
Documentation
Alignment: Values & North Star
Table of Contents
- Purpose
- When to Use This Skill
- What is Values & North Star Alignment?
- Workflow
- Common Patterns
- Guardrails
- Quick Reference
Purpose
Create clear, actionable alignment frameworks that give teams a shared North Star (direction), values (guardrails), and decision tenets (behavioral standards). This enables autonomous decision-making while maintaining coherence across the organization.
When to Use This Skill
- Starting new teams or organizations (defining identity)
- Scaling teams (maintaining culture as you grow)
- Resolving misalignment or conflicts (clarifying shared direction)
- Defining product/engineering/design principles
- Creating strategic clarity after pivots or changes
- Establishing decision-making frameworks
- Setting cultural norms and behavioral expectations
- Post-merger integration (aligning different cultures)
- Crisis response (re-centering on what matters)
- Onboarding leaders who need to understand team identity
Trigger phrases: "North Star", "team values", "mission", "vision", "principles", "guardrails", "what we stand for", "decision framework", "cultural alignment", "operating principles"
What is Values & North Star Alignment?
A framework with three layers:
- North Star: The aspirational direction - where are we going and why?
- Values/Guardrails: Core principles that constrain how we operate
- Decision Tenets/Behaviors: Concrete, observable behaviors that demonstrate values
Quick Example:
# Engineering Team Alignment
## North Star
Build systems that developers love to use and operators trust to run.
## Values
- **Simplicity**: Choose boring technology that works over exciting technology that might
- **Reliability**: Every service has SLOs and we honor them
- **Empathy**: Design for the developer experience, not just system performance
## Decision Tenets
When choosing between options:
✓ Pick the solution with fewer moving parts
✓ Choose managed services over self-hosted when quality is comparable
✓ Optimize for debuggability over micro-optimizations
✓ Document decisions (ADRs) for future context
## Behaviors (What This Looks Like)
- Code reviews comment on operational complexity, not just correctness
- We say no to features that compromise reliability
- Postmortems focus on learning, not blame
- Documentation is part of "done"
Workflow
Copy this checklist and track your progress:
Alignment Framework Progress:
- [ ] Step 1: Understand context
- [ ] Step 2: Choose framework
- [ ] Step 3: Develop alignment artifact
- [ ] Step 4: Validate quality
- [ ] Step 5: Deliver and socialize
Step 1: Understand context
Gather background: team/organization (size, stage, structure), current situation (new team, scaling, misalignment, crisis), trigger (why alignment needed NOW), stakeholders (who needs to align), hard decisions (where misalignment shows up), and existing artifacts (mission, values, culture statements). This ensures the framework addresses real needs.
Step 2: Choose framework
For new teams/startups (< 30 people, defining identity from scratch) → Use resources/template.md. For scaling organizations (existing values need refinement, multiple teams, need decision framework) → Study resources/methodology.md. To see examples → Review resources/examples/ (engineering-team.md, product-vision.md, company-values.md).
Step 3: Develop alignment artifact
Create alignment-values-north-star.md with: compelling North Star (1-2 sentences, aspirational but specific), 3-5 core values (specific to this team, not generic), decision tenets ("When X vs Y, we..."), observable behaviors (concrete examples), anti-patterns (optional - what we DON'T do), and context (optional - why these values). See Common Patterns for team-type specific guidance.
Step 4: Validate quality
Self-check using resources/evaluators/rubric_alignment_values_north_star.json. Verify: North Star is inspiring yet concrete, values are specific and distinctive, decision tenets guide real decisions, behaviors are observable/measurable, usable for decisions TODAY, trade-offs acknowledged, no contradictions, distinguishes this team from others. Minimum standard: Score ≥ 3.5 (aim for 4.5+ if organization-wide).
Step 5: Deliver and socialize
Present completed framework with rationale (why these values), examples of application in decisions, rollout/socialization approach (hiring, decision-making, onboarding, team meetings), and review cadence (typically annually). Ensure team can recall and apply key points.
Common Patterns
For technical teams:
- Focus on technical trade-offs (simplicity vs performance, speed vs quality)
- Make architectural principles explicit
- Include operational considerations
- Address technical debt philosophy
For product teams:
- Center on user/customer value
- Address feature prioritization philosophy
- Include quality bar and launch criteria
- Make product-market fit assumptions explicit
For company-wide values:
- Keep values aspirational but grounded
- Include specific behaviors (not just values)
- Address how values interact (what wins when they conflict?)
- Make hiring/firing implications clear
For crisis/change:
- Acknowledge what's changing
- Re-center on core that remains
- Be explicit about new priorities
- Include timeline for transition
Guardrails
Do:
- Make values specific and distinctive (not generic)
- Include concrete behaviors and examples
- Acknowledge trade-offs (what you're NOT optimizing for)
- Test values against real decisions
- Keep it concise (1-2 pages max)
- Make it memorable (people should be able to recall key points)
- Involve the team in creating it (not top-down)
Don't:
- Use corporate jargon or buzzwords
- Make it so generic it could apply to any company
- Create laundry list of every good quality
- Ignore tensions between values
- Make it purely aspirational (need concrete behaviors)
- Set it and forget it (values should evolve)
- Weaponize values to shut down dissent
Quick Reference
- Standard template:
resources/template.md - Scaling/complex cases:
resources/methodology.md - Examples:
resources/examples/engineering-team.md,resources/examples/product-vision.md,resources/examples/company-values.md - Quality rubric:
resources/evaluators/rubric_alignment_values_north_star.json
Output naming: alignment-values-north-star.md or {team-name}-alignment.md
Quick Install
/plugin add https://github.com/lyndonkl/claude/tree/main/alignment-values-north-starCopy and paste this command in Claude Code to install this skill
GitHub 仓库
Related Skills
sglang
MetaSGLang is a high-performance LLM serving framework that specializes in fast, structured generation for JSON, regex, and agentic workflows using its RadixAttention prefix caching. It delivers significantly faster inference, especially for tasks with repeated prefixes, making it ideal for complex, structured outputs and multi-turn conversations. Choose SGLang over alternatives like vLLM when you need constrained decoding or are building applications with extensive prefix sharing.
evaluating-llms-harness
TestingThis Claude Skill runs the lm-evaluation-harness to benchmark LLMs across 60+ standardized academic tasks like MMLU and GSM8K. It's designed for developers to compare model quality, track training progress, or report academic results. The tool supports various backends including HuggingFace and vLLM models.
llamaguard
OtherLlamaGuard is Meta's 7-8B parameter model for moderating LLM inputs and outputs across six safety categories like violence and hate speech. It offers 94-95% accuracy and can be deployed using vLLM, Hugging Face, or Amazon SageMaker. Use this skill to easily integrate content filtering and safety guardrails into your AI applications.
langchain
MetaLangChain is a framework for building LLM applications using agents, chains, and RAG pipelines. It supports multiple LLM providers, offers 500+ integrations, and includes features like tool calling and memory management. Use it for rapid prototyping and deploying production systems like chatbots, autonomous agents, and question-answering services.
