DOCSEARCHbyAlgolia
Submit Docs

pnp/copilot-prompts

repository·main·Indexed 21 days ago

https://github.com/pnp/copilot-prompts

A community-driven repository of sample prompts, agent instructions, and GitHub Copilot skills for Microsoft Copilot, GitHub Copilot, and Microsoft 365 Copilot. The collection includes prompt samples for direct use, agent instructions for Copilot Studio, and reusable skill files (SKILL.md) for multi-step tasks in GitHub Copilot. It also features a specialized Use Case Qualifier agent designed to evaluate AI solutions across five weighted dimensions and recommend Microsoft AI platforms.

Tokens
122.4K
Snippets
205
Records
443
Agent score
74%

What's inside pnp-copilot-prompts

  1. Overview of the Use Case Qualifier Agent

    main

    The Use Case Qualifier is a Microsoft 365 Copilot agent designed for partners, consultants, and AI program leads. It facilitates a structured qualification conversation to evaluate AI solutions before development begins.

    Key capabilities include:

    • Scoring use cases across five weighted dimensions.
    • Recommending the appropriate Microsoft AI platform (e.g., M365 Copilot, Copilot Studio, Azure AI Foundry).
    • Identifying top risks and success factors.
    • Generating a structured, shareable qualification report.

    The methodology is based on the Microsoft Jumpstart use case qualification approach used in enterprise AI programs.

  2. Overview of the Strategic Mindset Agent

    main
    The Strategic Mindset Agent is a coaching prompt designed to guide users through a five-round iterative process. Its primary goal is to help users develop a strategic mindset and avoid falling into purely tactical thinking patterns. It acts as a peer-level expert with 25 years of multi-domain experience, using a structured framework to identify goals and ambitions through iterative questioning.

  3. Overview of the Learning Path Architect agent

    main
    The Learning Path Architect is an intelligent AI agent designed to create personalized, adaptive learning journeys. It functions by analyzing a user's career goals, current skill sets, and learning preferences. Specifically, it can analyze professional development needs from Microsoft 365 data to curate custom learning paths that include specific resources, milestones, and progress tracking to accelerate career growth.

  4. Overview of the Personal Insurance Advisor Agent

    main

    The Personal Insurance Advisor Agent is an Agent Builder agent designed to help users understand insurance coverage, identify coverage gaps, and compare insurance quotes. It translates complex policy language into plain English and validates whether price differences in quotes reflect true coverage variations or structural weaknesses.

    Supported Coverage Types

    • Homeowners
    • Renters
    • Auto
    • Umbrella
    • Flood
    • Landlord
    • Life
    • Disability

    Key Capabilities

    • Coverage Summary: Explains policy types, carriers, premiums, limits, deductibles, endorsements, and exclusions.
    • Adequacy Analysis: Evaluates if coverage is adequate, insufficient, or excessive based on assets and risk profiles.
    • Policy Comparison: Normalizes multiple quotes to separate price differences from coverage differences.
    • Savings & Renewal Analysis: Calculates dollar and percentage changes between terms or quotes and classifies savings (e.g., "Clean savings" vs "Tradeoff savings").
    • Action Planning: Provides prioritized recommendations and specific questions to ask licensed agents.

  5. Overview of the Chief of Staff (Agent)

    main

    The Chief of Staff is a Microsoft 365 Copilot agent designed to help users regain control of their workload. It provides a structured view of emails, meetings, messages, and tasks, surfacing priorities, risks, and upcoming commitments. The agent acts as a productivity strategist that recommends actions (such as email replies, calendar changes, or task creation) but requires explicit user confirmation before executing any action.

    Key Capabilities:

    • Workload Triage: Summarizes missed communications and overdue tasks.
    • Priority Scoring: Ranks work items using a configurable weighting profile.
    • Stakeholder Sentiment Detection: Identifies urgency and frustration in communications.
    • Meeting Audits: Identifies low-value meetings and suggests rescheduling or declining.
    • Calendar Optimization: Creates protected focus blocks aligned with top priorities.

  6. Overview of the Communication Assistant agent

    main
    The Communication Assistant is an AI agent designed to help users communicate more considerately and avoid unintentionally hurting others. It evaluates text and images from diverse cultural, religious, and social perspectives to suggest more inclusive and thoughtful expressions. It acts as a 'growing partner' that values user creativity rather than forcing words into a bland mold.

  7. Overview of the Communication Assistant for a Kinder World

    main

    The Communication Assistant is an AI agent designed to embody Microsoft's 'Responsible AI' principles, specifically focusing on 'Inclusiveness,' 'Fairness,' and 'Reliability and Safety.'

    Unlike standard content safety filters (like Azure Content Safety) that focus on blocking critical issues (Sexual, Violent, Hate, Self-harm), this agent is a complementary tool designed to improve the quality of communication. It focuses on:

    • Nuanced Evaluation: Addressing 'gray zone' communication, such as cultural/religious nuances, psychosocial impacts, accessibility, and the perception of humor or satire.
    • Constructive Support: Providing specific suggestions on how to convey messages more warmly or resonate with diverse audiences without overriding user creativity.
    • Proactive Awareness: Encouraging users to consider diverse perspectives before communication issues arise, aiming to build smoother, more heartfelt interactions.

  9. Overview of the Prompt-of-the-Week PPTX Skill

    main

    The prompt-of-the-week skill automates the creation of a weekly PowerPoint slide from a GitHub sample URL. It functions like a 'mail merge': it fetches data from a GitHub repository (README and sample.json), extracts specific fields, and injects them into a pre-designed Prompt-K.pptx template using XML editing.

    Workflow Summary:

    1. Fetch & Parse: Retrieve the GitHub README and sample.json.
    2. Extract: Pull five specific data fields (Title, Prompt, Author, Tags, and Top Tile type).
    3. Inject: Copy the template and use XML string replacement to fill the slots.
    4. Swap Photo: Replace the template's default photo with the current week's author photo from the skill's author/ folder.
    5. QA: Convert to PDF/JPEG to visually verify the layout before final output.

  10. What is a GitHub Copilot Skill?

    main

    A Skill is a reusable instruction file (SKILL.md) that teaches GitHub Copilot how to perform a specific, multi-step, repeatable task within a codebase or workflow. Unlike a single natural-language Prompt or a system-level Agent Instruction, a Skill is a structured markdown file that Copilot reads at runtime to gain domain-specific expertise.

    Key Differences

    ConceptWhat it isWhere it runs
    PromptA single natural-language instruction the user typesMicrosoft 365 Copilot, Microsoft Copilot, or GitHub Copilot Chat
    Agent InstructionA system prompt pasted into Copilot Studio to create an agentMicrosoft Copilot Studio
    SkillA reusable instruction file that teaches GitHub Copilot a multi-step taskGitHub Copilot in VS Code (via .github/skills/ or workspace config)

  11. What is the Learning Path Architect Agent?

    main
    The Learning Path Architect is an AI career development coach designed to create personalized learning journeys. It analyzes professional context (such as Microsoft 365 data including work projects, communications, and document creation) to assess current skills, identify growth opportunities, and curate high-quality learning resources. Its primary mission is to align learning paths with specific career aspirations and connect theoretical learning to real-world application.

  12. What is the SPRINT Governance Agent?

    main

    The SPRINT Governance Agent is a productivity tool designed for AI/Copilot governance teams. It uses the SPRINT framework to transform scattered policy requirements into structured, trackable controls for AI agents, flows, or Copilot Studio deployments.

    The framework evaluates agents across six dimensions:

    • S - Security: Authentication, connector permissions, environment segregation, and least-privilege access.
    • P - Privacy by Design: Data minimization, DLP policy alignment, and sensitive data classification.
    • R - Responsible AI by Default: Content safety, grounding requirements, and human-in-the-loop checkpoints.
    • I - Identity: Entra Agent ID registration, service principal scoping, and clear ownership.
    • N - No Retrofit Rule: Ensuring governance is designed at build time rather than bolted on later.
    • T - Track and Measure: Audit logging, Purview integration, usage telemetry, and review cadences.