AI-Enablement-Academy

copilot-studio-builder

Build Microsoft Copilot Studio agents (the standalone low-code agent platform on Power Platform) as paste-ready deliverables for your organization or clients. Use when user says 'Copilot Studio', 'build a Studio agent', 'low-code agent', 'autonomous agent for [client]', 'Power Platform agent', 'Copilot Studio proposal/estimate'. Runs the full pipeline: client brief intake, use-case pattern selection, orchestration decision (generative vs classic), knowledge + tools spec, guardrails, test plan, deployment guide with a Copilot Credit licensing estimate. Boundary: for agents built inside M365 Copilot itself (declarative agents, agent builder) use m365-copilot-builder instead.

m365-copilot-builder

Build Microsoft 365 Copilot agents (declarative agents via Copilot agent builder) as paste-ready deliverables for your organization or clients. Use when user says 'build a Copilot agent', 'M365 Copilot agent', 'declarative agent', 'agent builder', 'Copilot for [client]', 'Copilot agent for [any client name]'. Runs the full pipeline: client brief intake, use-case pattern selection, instruction writing (8,000-char Copilot format), knowledge source spec, guardrails, test script, and a business-user setup guide. For the standalone Copilot Studio low-code platform (topics, autonomous agents, channels) use copilot-studio-builder instead.

custom-gpt-builder

Build OpenAI Custom GPTs (ChatGPT GPTs) as paste-ready deliverables for your organization or clients. Use when user says 'build a custom GPT', 'GPT for [client]', 'ChatGPT GPT', 'custom gpt creator', 'publish to the GPT Store', 'make a GPT'. Runs the full pipeline: client brief intake, use-case pattern selection, instruction writing (8,000-char GPT format with anti-prompt-extraction hardening), knowledge file spec, capabilities/Actions decisions, guardrails, test script, and a business-user setup guide with a workspace-admin split. For M365 Copilot declarative agents use m365-copilot-builder; for Copilot Studio low-code agents use copilot-studio-builder.

ai-opportunity-finder

Coaching skill that helps teams find where AI creates the most impact in their own processes. Use when user says 'find AI opportunities', 'where can we use AI', 'AI use cases for my business/team/ops', 'what should we automate', 'AI opportunity assessment', or 'identify processes for AI'. This is a coaching skill, not a consulting one - it asks questions instead of prescribing solutions, running Design Thinking, 5 Whys, First Principles, and Second Order Thinking to surface pain points and root causes, then IERF (Impact, Effort, Feasibility, Readiness) to prioritize - ending in a prioritized opportunity register the team discovered themselves.

Data & Documents Listed

boolean-search-string-writer

Write copy-pasteable Boolean search strings for sourcing recruiters across LinkedIn Recruiter, free LinkedIn, Google X-ray, GitHub, and ATS/CSE. Use this whenever the user is sourcing or building a candidate pipeline and needs search strings, Boolean strings, X-ray searches, or sourcing queries — even if they only say "find me people for this role," "help me search LinkedIn," "turn this job description into a search," or "my search returns nobody/too many." Ingests a job description, ICP, and/or talent-intelligence brief in any combination, interviews for the gaps, pushes back on over-narrow or biased strings, and produces several tuned strings (precise, broad, target-company, Google X-ray) with a widen/narrow tuning guide. Trigger it for sourcing-string work even when the user does not use the word "Boolean."

Data & Documents Listed

candidate-outreach-writer

Write short, personalized recruiting outreach messages to a specific candidate (LinkedIn InMail, email, or other channel) using a tree-of-thought loop — generate 5 genuinely different drafts, grade each against a message rubric, improve them, regrade, and ship the single best one with its scorecard. Use this whenever a recruiter or sourcer wants to reach out to, message, contact, or "DM" a named candidate, write an InMail or recruiting cold email, personalize outreach, improve a draft that's getting ignored, or turn a candidate profile (plus an ICP and/or job description, if they have them) into a message — even if they just paste a LinkedIn profile and say "write something to get them on a call." This is the outreach stage of the recruiting pipeline (tool 6 of 6).

Web & Frontend Listed

hiring-packet-builder

Turn a candidate's resume plus interview transcript or screen notes into a one-page hiring packet (candidate write-up / submittal) to send to the hiring manager — advocating with evidence, mapped to the role's must-haves, honest about risks, and EEOC-compliant. Use this whenever a recruiter or talent acquisition partner has screened or interviewed someone and wants to present them, even if they just say "write up this candidate," "build a submittal," "draft a candidate profile for the hiring manager," "summarize my interview notes into a write-up," "make the case for this person," or paste a resume and screen notes and ask what to send the HM. Use it after screening/interviewing and before the hiring manager reviews the candidate.

Data & Documents Listed

ideal-candidate-profile-creator

Turn hiring-manager intake notes (or a pasted intake-call transcript) plus a job description into 2–4 distinct Ideal Candidate Profiles that drive sourcing and talent-intelligence research. Use this skill whenever a recruiter or talent acquisition partner has run (or is about to run) a hiring-manager intake and wants to define WHO does the role well — even if they just say "build candidate profiles," "create an ICP," "who should I be sourcing," "turn this intake into personas," "help me figure out who to go find for this role," or paste intake notes + a JD and ask what to do with them. Use it after the JD Clarifier and the intake, and before sourcing, Boolean search, screening, or talent-market research.

job-description-clarifier

Analyze a job description before the hiring manager intake and produce a gap/risk analysis, a prioritized hiring-manager intake question set, and a partial Role Brief. Use this whenever a recruiter or talent acquisition partner shares a job description, job posting, job ad, JD, or role spec and is preparing to meet a hiring manager — even if they just say "help me prep for an intake," "find the gaps in this role," "what should I ask the hiring manager," "this JD is vague," or paste a job posting and ask what to do with it. Use it before writing an ICP, sourcing, or screening — this is the front of the recruiting pipeline.

Web & Frontend Listed

recruiter-screen-designer

Design a structured candidate-screen guide for a recruiter — the exact questions to ask THIS candidate to test "fit for match" (does the candidate meet this role's requirements now) and "fit for scale" (can the candidate grow with the company), plus a 0–5 scoring rubric and a ~30-minute screen flow. Use this whenever a recruiter or talent acquisition partner is preparing to phone-screen or first-round-screen a candidate and shares any of — a resume/CV, an ICP/candidate profile, a job description, hiring-manager intake notes, or a role brief — even if they just say "help me prep for this screen," "what should I ask this candidate," "build me a phone screen," "I'm screening someone tomorrow," "design screening questions," or paste a resume and ask what to probe. Use it before the recruiter runs the screen and before the Hiring Packet Builder.

talent-intelligence-brief-writer

Write a self-contained talent-intelligence research brief that can be handed to an analyst with zero prior context. Use this skill whenever the user wants to commission market research, a talent map, a market map, a labor-market or supply/demand study, a competitor headcount/attrition scan, a compensation benchmark, a location/build-vs-buy analysis, or DEI availability data for a hiring decision — even if they just say "find out what the market looks like" or "brief our analyst." Turns a JD + ICP + hiring-manager intake (or whatever the user has) into a precise, decision-anchored research request. Refuses to write a vague "tell me about the market" topic brief.