personamanagmentlayer__pcl/stdlib

PCL Standard Library (stdlib)

Version: 3.1.0 Status: ✅ Production Ready Total Skills: 191 Expert-Level Skills

---

🎯 What is the Standard Library?

The PCL Standard Library is a comprehensive collection of 191 expert-level skills for AI agent personas, providing deep domain expertise across programming languages, frameworks, cloud platforms, data systems, security, and industry verticals.

Each skill represents expert-level mastery in a specific domain with:

• Core Concepts: Fundamental principles and architecture
• Code Examples: Production-ready implementations (250-800+ lines)
• Best Practices: Industry standards and proven patterns
• Anti-Patterns: Common mistakes to avoid
• Resources: Documentation, tools, and learning materials

📊 Library Statistics

• Total Skills: 191 expert-level skills
• Categories: 15 major domains
• Total Content: ~114,000 lines of expert knowledge (48,000 in skill entry points, 66,000 in on-demand reference documents)
• Coverage: Languages, Frameworks, Cloud, Data, Security, DevOps, Industries, and more

🗂️ Skill Categories

Category	Count	Examples
Domains	58	Healthcare, Finance, LegalTech, Manufacturing, Energy, AgTech, Web3
Workflows	4	TDD, refactoring, debugging, code review — procedures, not knowledge
Languages	23	Python, TypeScript, Rust, Go, Java, Kotlin, Scala, Haskell, Julia, Zig
DevOps	16	Kubernetes, Docker, Terraform, ArgoCD, Prometheus, Grafana, Istio
Frameworks	15	React, Vue, Angular, Next.js, Django, Spring Boot, Flutter, Tauri
Data & Analytics	17	Snowflake, Databricks, Airflow, dbt, Tableau, Power BI, Kafka
Security	12	Penetration Testing, Zero Trust, GDPR, SOC2, Cryptography
Tools	12	Git, Slack, Teams, Discord, WebRTC, Video Streaming, Code Review
QA & Testing	7	Playwright, Cypress, Jest, Selenium, Load Testing, Chaos Engineering
API & Integration	6	REST, GraphQL, gRPC, Microservices, API Design
Professional	5	Banking, Legal, Accounting, FinOps, Standards
AI & ML	7	Machine Learning, AI Architecture, Data Science
Cloud Platforms	4	AWS, Azure, GCP, Cloudflare
Scientific	3	Quantum Computing, Bioinformatics, Research
Design	2	System & UX Design

See SKILLS_INVENTORY.md for the complete skill list.

🚀 Quick Start

1. Browse Available Skills

# List all skills
find stdlib -name SKILL.md

# Search for a specific domain
find stdlib -type d -name "*react*"
find stdlib -type d -name "*kubernetes*"

# Machine-readable index
cat stdlib/catalog/skill-index.json

2. Load Skills into a Persona

# persona.pcl
persona "FullStackDeveloper" {
  version = "1.0.0"

  # Load skills from stdlib
  skills = [
    "stdlib/languages/typescript-expert",
    "stdlib/frameworks/react-expert",
    "stdlib/frameworks/nextjs-expert",
    "stdlib/data/postgresql-expert",
    "stdlib/devops/docker-expert"
  ]

  constraints {
    max_tokens = 8000
    allowed_tools = ["Read", "Write", "Edit", "Bash"]
  }
}

3. Use Skills in Multi-Agent Systems

# multi-agent-system.pcl
system "DataPlatform" {

agent "DataEngineer" {
skills = [
"stdlib/data/airflow-expert",
"stdlib/data/dbt-expert",
"stdlib/data/snowflake-expert",
"stdlib/languages/python-expert"
]
role = "Build and maintain data pipelines"
}

agent "DataAnalyst" {
skills = [
"stdlib/data/tableau-expert",
"stdlib/data/powerbi-expert",
"stdlib/data/sql-expert"
]
role = "Create dashboards and analytics"
}

agent "MLEngineer" {
skills = [
"stdlib/ai/ml-expert",
"stdlib/data/databricks-expert",
"stdlib/languages/python-expert"
]
role = "Train and deploy ML models"
}
}

📖 Skill Format

Each skill follows the PCL Agent Skills Specification:

---
name: skill-name
version: 1.0.0
description: Brief description of the skill
category: domain
tags: [tag1, tag2, tag3]
allowed-tools:
  - Read
  - Write
  - Edit
  - Bash(command:*)
---

# Skill Title

Brief introduction to the skill domain.

## Core Concepts

### Concept 1

Explanation...

## Code Examples

\`\`\`language
// Production-ready code example
// 50-200 lines demonstrating real-world usage
\`\`\`

## Best Practices

- Practice 1
- Practice 2

## Anti-Patterns

❌ Anti-pattern 1
❌ Anti-pattern 2

## Resources

- Documentation: https://...
- Tools: https://...

🎭 Example Personas

Full-Stack Web Developer

persona "FullStackDev" {
  description = "Expert full-stack web developer"

  skills = [
    # Frontend
    "stdlib/languages/typescript-expert",
    "stdlib/frameworks/react-expert",
    "stdlib/frameworks/nextjs-expert",

    # Backend
    "stdlib/languages/nodejs-expert",
    "stdlib/frameworks/fastapi-expert",
    "stdlib/api/graphql-expert",

    # Database
    "stdlib/data/postgresql-expert",
    "stdlib/data/redis-expert",

    # DevOps
    "stdlib/devops/docker-expert",
    "stdlib/devops/kubernetes-expert",
    "stdlib/cloud/aws-expert"
  ]

  constraints {
    governance_level = "high"
    risk_classification = "medium"
  }
}

Cloud Security Architect

persona "SecurityArchitect" {
  description = "Expert in cloud security and compliance"

  skills = [
    # Security
    "stdlib/security/zero-trust-expert",
    "stdlib/security/cryptography-expert",
    "stdlib/security/penetration-testing-expert",

    # Compliance
    "stdlib/security/gdpr-expert",
    "stdlib/security/soc2-expert",

    # Cloud
    "stdlib/cloud/aws-expert",
    "stdlib/cloud/azure-expert",
    "stdlib/devops/kubernetes-expert",
    "stdlib/devops/istio-expert"
  ]

  constraints {
    governance_level = "critical"
    risk_classification = "high"
    audit_required = true
  }
}

Data Platform Engineer

persona "DataEngineer" {
  description = "Expert in modern data platforms"

  skills = [
    # Languages
    "stdlib/languages/python-expert",
    "stdlib/languages/scala-expert",

    # Data Platforms
    "stdlib/data/snowflake-expert",
    "stdlib/data/databricks-expert",

    # Orchestration
    "stdlib/data/airflow-expert",
    "stdlib/data/dbt-expert",

    # Streaming
    "stdlib/data/kafka-expert",

    # Cloud
    "stdlib/cloud/aws-expert",
    "stdlib/devops/terraform-expert"
  ]

  constraints {
    governance_level = "high"
    data_classification = ["pii", "confidential"]
  }
}

🔍 Skill Discovery

By Technology Stack

MERN Stack:

• mongodb-expert
• nodejs-expert
• react-expert
• typescript-expert

Data Science Stack:

• python-expert
• data-science-expert
• ml-expert
• databricks-expert

Cloud Native Stack:

• kubernetes-expert
• istio-expert
• prometheus-expert
• grafana-expert
• argocd-expert

Mobile Development:

• flutter-expert
• react-native-expert
• ios-expert
• android-expert

By Industry

Healthcare: healthcare, healthtech-expert, pharmaceutical-expert Finance: banking-expert, fintech-expert, trading-expert Legal: lawyer-expert, legal-tech, legaltech-expert Manufacturing: manufacturing-expert, iot-expert Agriculture: farming, agtech-expert

🛡️ Governance & Security

Skills integrate with PCL's governance framework:

Risk Classification

skill "penetration-testing-expert" {
  risk_level = "high"          # Can execute security tests
  requires_approval = true      # Human approval needed
  audit_all_actions = true      # Log everything
}

skill "react-expert" {
  risk_level = "low"           # UI development only
  requires_approval = false
}

Tool Permissions

# Skill declares what tools it needs
allowed-tools:
  - Read # Can read files
  - Write # Can write files
  - Edit # Can edit files
  - Bash(npm:*, yarn:*) # Can run npm/yarn only
  - Bash(kubectl:get:*) # Can only run kubectl get

Compliance Alignment

Skills support compliance frameworks:

• ISO 42001 - AI risk management
• ISO 27001 - Information security
• GDPR - Data protection (gdpr-expert skill)
• SOC 2 - Security controls (soc2-expert skill)
• OWASP - Security best practices

📚 Skill Composition Patterns

Layered Architecture

persona "EnterpriseBackend" {
  skills = [
    # Layer 1: Language
    "stdlib/languages/java-expert",

    # Layer 2: Framework
    "stdlib/frameworks/spring-boot-expert",

    # Layer 3: Data
    "stdlib/data/postgresql-expert",
    "stdlib/data/redis-expert",

    # Layer 4: Integration
    "stdlib/api/microservices-expert",
    "stdlib/api/graphql-expert",

    # Layer 5: Infrastructure
    "stdlib/devops/kubernetes-expert",
    "stdlib/devops/istio-expert",
    "stdlib/cloud/aws-expert"
  ]
}

Cross-Functional Teams

system "ProductDevelopment" {

persona "Frontend" {
skills = ["react-expert", "typescript-expert", "cypress-expert"]
}

persona "Backend" {
skills = ["go-expert", "postgresql-expert", "microservices-expert"]
}

persona "DevOps" {
skills = ["kubernetes-expert", "terraform-expert", "prometheus-expert"]
}

persona "Security" {
skills = ["penetration-testing-expert", "zero-trust-expert"]
}
}

📊 Documentation

• CHANGELOG.md - Per-skill versioning and migration notes
• SKILLS_INVENTORY.md - Complete skill list with sizes and reference files
• DIRECTORY_STRUCTURE.md - File organization and structure
• catalog/ - Machine-readable index (JSON + YAML)

Both documents are generated; rerun python scripts/generate-skill-inventory.py after changing the tree rather than editing them.

🔧 Development & Contributing

Creating Custom Skills

1. Copy an existing skill as template:

cp -r stdlib/languages/python-expert stdlib/languages/my-skill

2. Edit SKILL.md with your expertise

3. Validate the skill:

python scripts/validate-skills.py

Maintenance commands

python scripts/validate-skills.py           # conformance gate (must pass)
python scripts/split-skill-references.py    # keep SKILL.md under 500 lines
python scripts/generate-skill-catalog.py    # rebuild catalog/
python scripts/generate-skill-inventory.py  # rebuild inventory + structure docs

See CONTRIBUTING.md for contribution guidelines.

📈 What's New in v3.1.0

17 new skills, taking the library from 174 to 191 and adding a fifteenth category. Chosen by cross-referencing coverage gaps against observed usage — developer survey data and the install rankings of public agent-skill registries.

• workflows/ (new category, 4 skills) — tdd-workflow, refactoring-workflow, debugging-workflow, code-review-workflow. Procedures rather than knowledge: the shape that dominates real agent-skill usage and that the library had none of.
• Applied AI (3) — llm-engineering-expert, rag-expert, agent-engineering-expert. The largest coverage gap: the library mentioned LLMs in tags but had no skill on prompting, retrieval or tool use.
• Capabilities (2) — document-processing-expert (PDF, DOCX, XLSX, PPTX in one skill), browser-automation-expert (distinct from playwright-expert, which is for testing your own application).
• Data (3) — mysql-expert (the second most used database, previously absent), analytical-databases-expert, stream-processing-expert.
• Security (3) — secrets-management-expert, supply-chain-security-expert (OWASP A03 2025), identity-access-expert (OWASP A01).
• Interface (1) — accessibility-expert. The design/ category held only software architecture; WCAG conformance had no home.
• Meta (1) — skill-router, for finding and composing skills across 191.

Grouping was deliberate: five vector stores live inside rag-expert, four document formats inside document-processing-expert. Roughly fifty topics are covered by seventeen skills, with per-tool detail in references/.

📈 What's New in v3.0.0

Agent Skills v1.0 conformance. Every skill now passes scripts/validate-skills.py. This release has breaking changes — one skill removed, five moved, seven with capabilities withdrawn. See CHANGELOG.md for the migration notes.

• Uniform layout — all 174 skills live at stdlib/<category>/<name>/SKILL.md; 71 flat files were converted, and 5 directories renamed to match their declared name:.
• Repaired metadata — 20 skills had an unparseable frontmatter block (an H1 preceded it, and the key was skill_id: instead of name:), so tooling saw no metadata at all; 21 more were missing version:.
• Discoverable descriptions — every description states an explicit Use when ... trigger naming the keywords and tasks that select the skill.
• Progressive disclosure — 127 oversized skills were split, moving bulk material into references/ read on demand. A SKILL.md now averages 277 lines instead of 644, so activating a skill costs a fraction of the context it did.
• Deduplicated — the two overlapping legal-technology skills were merged into legaltech-expert.

📈 What's New in v2.0.0

Major Expansion: 100 → 173 skills (+73 skills)

New Categories Added:

• Modern Languages: Zig, Nim, Julia, Haskell, Clojure
• Modern Frameworks: Vue, Angular, Svelte, Next.js, Remix, Flutter, Tauri
• Emerging Tech: Web3, Metaverse, Edge Computing, Serverless, WebAssembly, Robotics, 5G
• Data Platforms: Snowflake, Databricks, Airflow, dbt, Looker, Tableau, Power BI
• Security & Compliance: Penetration Testing, Zero Trust, GDPR, SOC2, Cryptography
• Cloud Native: Istio, Helm, ArgoCD, Prometheus, Grafana, Linkerd
• Mobile & Desktop: Flutter, React Native, iOS, Android, Electron, Tauri
• Industry Specializations: Pharmaceutical, Biotech, FinTech, RegTech, PropTech, CleanTech
• Enterprise Systems: SAP, Salesforce, ServiceNow, Workday, Oracle, Microsoft 365
• Testing & QA: Playwright, Cypress, Jest, Selenium, Load Testing, Chaos Engineering
• Communication: Slack, Teams, Discord, WebRTC, Video Streaming

📝 License

Part of the PCL project. See LICENSE for details.

🔗 Related Documentation

• PCL Specification
• Persona Guide
• Governance Framework
• Security Model

---

The PCL Standard Library - Governed AI expertise at scale 🚀

Last Updated: 2026-01-19 Maintainer: PCL Team Status: ✅ Production Ready