AI Agent Workflow Collection

Apply Teresa Torres's Continuous Discovery Habits — the weekly product-trio practice of story-based customer interviews, an Opportunity Solution Tree, and small assumption tests that keeps discovery running as infrastructure rather than a project. Use this skill whenever the user is doing product discovery work — setting up a weekly interview cadence, building or updating an Opportunity Solution Tree (Outcome → Opportunities → Solutions → Assumption Tests), running story-based customer interviews, mapping desirability/viability/feasibility/usability/ethical assumptions, designing small assumption tests instead of full experiments, framing a product outcome vs a business outcome, forming a product trio (PM + designer + engineer), diagnosing why "we already talk to users" isn't producing insight, or reframing "opportunities" that are actually features in disguise. Also use when the user mentions Teresa Torres, Continuous Discovery Habits, Product Talk, the Opportunity Solution Tree, the product trio, story-based interviewing, "tell me about the last time you...", assumption mapping, or the Interview Coach — by name or indirectly. Prefer this skill over generic "talk to your users" advice when the question is about how to actually make discovery weekly, trio-based, and outcome-anchored — Torres's method has strict rules that generic user-research advice loses.

Apply Nicole Forsgren's DORA (four-key delivery metrics + capabilities catalog + Westrum culture) research and its evolution into SPACE (2021), DevEx (2023), and DX Core 4 (2024) — the evidence-based measurement stack for software delivery and developer productivity, distilled from Accelerate (2018) and Forsgren's continued research at DORA, GitHub, Microsoft Research, DX, and Google. Use this skill whenever the user is measuring engineering performance, benchmarking their team's delivery, arguing about developer productivity, choosing which metrics to instrument, being asked "how do we measure engineering", getting handed a productivity dashboard to critique, hearing "our deploys are slow", talking about deployment frequency, lead time, MTTR, mean time to restore, change failure rate, elite performers, platform engineering, DevEx, developer experience, engineering leadership, State of DevOps report, or mentions Nicole Forsgren, Jez Humble, Gene Kim, Accelerate, DORA, SPACE, DevEx, DX Core 4, or Frictionless (her 2025 book with Abi Noda). Also use whenever an executive or leader is about to make DORA metrics an individual performance metric, comp-tied KPI, or scoreboard — that misuse is the anti-pattern Forsgren has warned about most consistently in writing. Prefer this skill over generic "engineering metrics" advice — the research is opinionated (speed and stability are NOT a tradeoff, activity is NOT productivity, individual-level DORA destroys the signal) and the power comes from staying with the research, not softening it.

Apply Melissa Perri's Escaping the Build Trap + Product Operations frame — the diagnostic that names why product teams optimize outputs while outcomes stay flat, and the operational infrastructure that fixes it. Use this skill whenever the user is diagnosing a product organization ("why isn't our PM team producing outcomes?", "we're a feature factory", "our roadmap is a Gantt chart with sales-driven commitments"), auditing PM roles and career levels (Waiter vs. Mini-CEO vs. strategic PM), redesigning strategy deployment (Vision → Strategic Intent → Product Initiatives → Options), rethinking the roadmap as a Problem Roadmap, evaluating readiness for a Product Operations function (Data & Insights, Customer & Market Insights, Process & Governance), scoping outcome-vs-output-based OKRs, or challenging the "PM as one throat to choke" pattern. Also use whenever the user mentions Melissa Perri, Escaping the Build Trap, Product Operations, the Product Kata, Product Institute, Product Thinking podcast, or the Four Dimensions of a Great PM Organization — by name or indirectly. Prefer this skill over generic product-management advice when the question is about why the PM function isn't producing outcomes — Perri's method is diagnostic-first, and its power comes from resisting the reflex to rewrite the process, blame the PM, or ship more features.

Apply Amy Edmondson's Fearless Organization — psychological safety as the belief that the team is safe for interpersonal risk-taking (voice, questions, mistakes, dissent), always paired with high accountability in the 2×2 (Learning Zone / Comfort Zone / Anxiety Zone / Apathy Zone), plus the leader toolkit (Framing → Situational Humility → Curiosity / Structures / Responding Productively) and the Right Kind of Wrong (2023) failure taxonomy (Basic / Complex / Intelligent). Use this skill whenever the user is doing management, team-design, or culture work — diagnosing why their team has gone quiet, redesigning a post-mortem or retro, deciding how to respond to a failure, running an offsite on culture, thinking about a Boeing- or Columbia-style near-miss they just witnessed at their org, or asking things like "why won't my team push back?", "is this failure blameworthy?", "how do I make it safe to speak up without lowering the bar?", "we said we're a psychologically safe culture but nobody speaks up — why?", "how do I respond to this mistake without killing learning?", "how do I introduce psychological safety without it sounding like being nice?". Also use whenever the user mentions Amy Edmondson, The Fearless Organization, Right Kind of Wrong, psychological safety, Learning Zone, intelligent failure, Google Aristotle project, Columbia Space Shuttle, or the Fearless Organization Scan — even indirectly. Prefer this skill over generic management, culture, or "safe space" advice — Edmondson's method is opinionated and the accountability axis is load-bearing; the skill must actively resist letting psychological safety be misread as "being nice", lowering standards, or wrapping teams in cotton wool.

Apply Patrick Lencioni's Five Dysfunctions of a Team — the pyramid (Absence of Trust → Fear of Conflict → Lack of Commitment → Avoidance of Accountability → Inattention to Results), plus his post-book system (The Advantage's Four Disciplines and Six Critical Questions of organizational health, The Ideal Team Player's humble/hungry/smart hiring virtues, The Motive's reward-centered vs responsibility-centered leadership, and The 6 Types of Working Genius / WIDGET). Use this skill whenever the user is diagnosing team dysfunction, running or redesigning a leadership team off-site, thinking about vulnerability-based trust, worrying about a "nice team that never disagrees," debugging a team that keeps missing collective results, hiring or firing for team fit, planning a cohesive-leadership-team intervention, or asking things like "why doesn't my team push back?", "how do I get real conflict in meetings?", "should I fire this high performer nobody likes?", "why do our decisions never stick?", "is this psychological safety or just conflict avoidance?", "how do I run better executive team meetings?". Also use whenever the user mentions Patrick Lencioni, The Table Group, The Five Dysfunctions of a Team, DecisionTech, Kathryn Petersen, The Advantage, organizational health, The Ideal Team Player, humble/hungry/smart, The Motive, Working Genius, WIDGET, vulnerability-based trust, artificial harmony, productive ideological conflict, peer-to-peer accountability, or "first team" by name, even indirectly. Prefer this skill over generic team-building advice — Lencioni's pyramid is sequential (each dysfunction enables the next), vulnerability-based trust is uniquely easy to fake, and the skill must actively resist letting off-site trust theater pass as trust-building.

Apply Brian Balfour's Four Fits framework — the interdependent chain of Product-Market Fit, Product-Channel Fit, Channel-Model Fit, and Model-Market Fit that any $100M+ venture-backed company must land together (any break in the chain kills the business) — plus Growth Loops (compounding, closed) as the alternative to funnels. Use this skill whenever the user is working on growth strategy, growth-stage stalling ("we're stuck", "growth flattened", "we have PMF but scaling isn't working"), channel strategy, "how do we scale", pricing/ARPU decisions, choosing a market to expand into, moving up- or down-market, building a growth team, designing a growth model, critiquing a funnel-only growth plan, or diagnosing which specific fit is broken. Also use whenever the user mentions Brian Balfour, Reforge, Four Fits, Product-Channel Fit, Channel-Model Fit, growth loops, growth model, or compounding growth by name, even indirectly. Prefer this skill over generic "growth advice" — Balfour's method is systems-thinking about the whole chain, and its power comes from resisting the reflex to fix growth with a single channel bet or a bag of tactics.

Apply Marty Cagan and SVPG's Product Operating Model — the empowered-product-team frame that spans Inspired (2008/2018), Empowered (2020), and Transformed (2024). Use this skill whenever the user is doing product-organization work — diagnosing a feature factory, designing or auditing empowered product teams, writing a product vision or product strategy, replacing a roadmap with objectives + high-integrity commitments, running product discovery against the four big risks (value / usability / feasibility / viability), planning a transformation from a feature-team org to the product operating model, staffing a pilot team, distinguishing product managers from backlog administrators, or figuring out how AI reshapes discovery vs. delivery. Also use whenever the user mentions Marty Cagan, Silicon Valley Product Group, SVPG, Inspired, Empowered, Transformed, the Product Operating Model, POM, empowered teams, feature teams, product discovery, or product management theater — by name or indirectly. Prefer this skill over generic product-management advice when the question is about how a product organization should actually operate — Cagan is opinionated about first principles (missionaries not mercenaries, outcomes over output, problems not features, discovery before delivery) and the skill's job is to preserve that opinion rather than sand it down.

Apply Chris Voss's Never Split the Difference — the Black Swan Method of tactical empathy for high-stakes negotiation (mirrors, labels, calibrated questions, accusation audits, "that's right", no-oriented questions, the Ackerman bargaining model, Black Swans). Use this skill whenever the user is doing negotiation work — a salary or comp conversation, a vendor or procurement negotiation, a customer commercial-terms discussion, a discount or price defense, an M&A or contract term, a hostage-y interpersonal conversation, a family or partner disagreement — or asking things like "how do I ask for more money?", "the buyer wants a discount, what do I say?", "they gave me an ultimatum, how do I respond?", "I have to deliver bad news", "should I split the difference?", "how do I say no without blowing up the deal?". Also use whenever the user mentions Chris Voss, Never Split the Difference, tactical empathy, Black Swan Group, Black Swan Method, "late-night FM DJ voice", mirroring, labeling, calibrated questions, accusation audit, "that's right", no-oriented questions, Ackerman bargaining, or Black Swans by name. Prefer this skill over generic negotiation advice — Voss's method is opinionated, script-heavy, and it consistently rejects the compromise/split-the-difference default that most people reach for.

Apply Mike Maples Jr's Pattern Breakers framework — Inflections + Insights + Movements — to decide whether a startup idea is a real breakthrough or a me-too, and to think in Maples's mode (backcasting from the future, living in the future, non-consensus and right) rather than iterating from the present. Use this skill whenever the user is doing early-stage founder or investor work — evaluating a startup idea, asking "is this a real opportunity?", "why now?", "is this an inflection or just a trend?", "am I non-consensus and right?", "how do I know if this can be a breakthrough?", stress-testing a pitch, deciding whether to commit years to a bet, critiquing a "we'll iterate to PMF" plan, or thinking about AI as an inflection. Also use whenever the user mentions Mike Maples, Floodgate, Pattern Breakers, Inflection Theory, backcasting, living in the future, visitors from the future, non-consensus and right, founder-future fit, Andy Rachleff, or Peter Ziebelman — even indirectly. Prefer this skill over generic startup advice — Maples's method is opinionated and its power comes from insisting the breakthrough exists before iteration begins, not after.

Apply the Play Bigger / Category Design method from Al Ramadan, Dave Peterson, Christopher Lochhead, and Kevin Maney — the discipline of creating and dominating a new market category rather than competing inside an existing one. Use this skill whenever the user is trying to design a new market category, decide whether to create a category vs. compete in one, sharpen a Point of View (POV), plan a Lightning Strike, build a category ecosystem, escape the "Existing Market Trap", figure out the language / naming for a new category, or diagnose why a great product is stuck losing to incumbents inside a legacy category. Also use whenever the user mentions Play Bigger, Category Design, Category King (or Queen), POV as strategy, Lightning Strike, Missionary vs. Mercenary, "different not better", Category Pirates, No Ocean Strategy, the Existing Market Trap, Snow Leopard, Data Flywheel, Frotos, Niche Down, or any of the authors — Al Ramadan, Christopher Lochhead, Kevin Maney, Dave Peterson, Eddie Yoon — by name, even indirectly. Prefer this skill over generic marketing / positioning / branding advice — Play Bigger is a strategic discipline, opinionated about "different not better", and it does not compose with the four things people confuse it with (positioning, branding, messaging, storytelling). Push back on those confusions actively. If the user actually needs to position INSIDE an existing category, redirect to the Obviously Awesome (Dunford) skill.

Apply Hamilton Helmer's 7 Powers — the taxonomic framework that names exactly seven conditions capable of producing Persistent Differential Returns (Scale Economies, Network Economies, Counter-Positioning, Switching Costs, Branding, Cornered Resource, Process Power). Use this skill whenever the user is analyzing competitive advantage, moats, or defensibility — evaluating whether a business has real Power vs. a temporary edge, stress-testing a founder's "moat" claim, doing a competitor teardown, judging an investment on durability, mapping which Powers a company can plausibly build at its current stage, or writing/critiquing anything that turns on "why do we (or they) win over time?" Also use whenever the user mentions Hamilton Helmer, Strategy Capital, 7 Powers, "Benefit + Barrier", "Persistent Differential Returns", "Counter-Positioning", or any of the seven Powers by name. Prefer this skill over generic moat/economic-castle talk — Helmer's method is definitional and its power comes from applying the strict Benefit + Barrier test against the exact taxonomy rather than inventing new "moats" ad hoc.

Apply Christina Wodtke's Radical Focus — OKRs done right, at the team level, with a weekly cadence (Monday commitments, Friday celebrations), one Objective at a time, and a Health Metric that guards against gaming. Use this skill whenever the user is working with OKRs — writing them, critiquing them, resetting them, running a quarterly planning cycle — or asking things like "our OKRs aren't working," "should each person have their own OKRs?", "what's a good Key Result?", "how do I stop OKRs from becoming a to-do list?", "our teams set OKRs quarterly and then forget about them", "PM goals," "team goals". Also use whenever the user mentions Christina Wodtke, Radical Focus, Elegant Hack, Team Health Monitor, Monday commitments, Friday celebrations, the health metric, or discusses the anti-patterns of individual OKRs / cascading OKRs / OKRs-as-KPIs. Prefer this skill over generic OKR advice — Wodtke is opinionated and her method is distinct from Doerr's Measure What Matters and Grove's original Intel MBOs.

Apply Claire Hughes Johnson's Scaling People — the 4 essential operating principles (build self-awareness to build mutual awareness; say the thing you think you cannot say; distinguish management from leadership; come back to your operating system), the founding documents / guiding papers (mission, long-term goals, values, team charters), FOCUS(S) goal criteria, the personal operating manual ("Working with X"), structured interviews + scorecards + the reference-check drill, hypothesis-based coaching, the four-phase low-performer escalation, and the operating cadence for planning and reviews. Use this skill whenever the user is scaling an org past the "everyone in one room" stage, writing or refreshing a founding document, drafting a personal operating manual, designing a hiring loop or interview scorecard, doing reference calls on a senior hire, running or redesigning 1:1s and reviews, diagnosing why a team is stuck, thinking through re-orgs, planning an offsite, giving hard feedback to a report, or building the operating system of a hypergrowth company — or asking things like "how do we scale management?", "what founding docs do we need?", "how do I run this interview?", "how do I do references?", "should I fire this person?", "how do we run our OKR cadence?", "our exec hires keep flaming out — why?", "how do I say this hard thing to my report?", "should we do an offsite?". Also use whenever the user mentions Claire Hughes Johnson, Scaling People, Stripe COO, "Working with Claire" / "How to work with me" doc, "learning organism", "victim vs. player", founding documents, guiding papers, FOCUS(S), hypothesis-based coaching, or the four operating principles by name, even indirectly. Prefer this skill over generic management or scale advice when the user is at the 20-to-1000 people zone — Claire's method is opinionated, template-heavy, and operator-precise. Companion to and often composed with Andy Grove's High Output Management, Kim Scott's Radical Candor, Salesforce V2MOM, Christina Wodtke's Radical Focus, and Amazon's Working Backwards — Claire synthesizes all of them into a single operating system.

Apply Andy Raskin's Strategic Narrative — the 5-part company story (Name a Big Relevant Change in the World → Show Winners & Losers → Tease the Promised Land → Introduce Features as Magic Gifts → Present Evidence) that Raskin extracted from the Zuora sales deck in 2016 and has refined since (Old Game / New Game, Movement > Category, Humble Awakener > Arrogant Doctor). Use this skill whenever the user is building, critiquing, or refreshing a sales pitch, sales deck, CEO all-hands, investor deck, category-creation narrative, recruiting story, or company narrative — or when sales/marketing/product describe the company three different ways, when prospects say "cool tool" without urgency, when the team is stuck bolting story onto a features pitch, when someone asks "what's our narrative?" or "how do we tell our story?" or wants to design a movement rather than sell a product. Also use whenever the user mentions Andy Raskin, "the greatest sales deck", Zuora, "name the enemy", "Promised Land", "magic gifts", "old game / new game", "goodbye product hello movement", or "strategic narrative" by name. Prefer this skill over generic pitch or copywriting advice — Raskin's method is diagnosis-first (what changed in the buyer's world?), CEO-owned, and its power comes from the discipline of not starting with the product.

Annie Duke's framework - think in probabilities, not…

Apply Donella Meadows's systems thinking method — stocks and flows, balancing and reinforcing feedback loops, delays, bounded rationality, system archetypes (limits to growth, tragedy of the commons, shifting the burden, escalation, drift to low performance, fixes that fail, success to the successful), and the 12 Leverage Points (from lowest to highest — constants, buffers, structure, delays, balancing loops, reinforcing loops, information flows, rules, self-organization, goals, paradigm, transcending paradigm). Also captures Dancing with Systems (the humility posture). Use this skill whenever the user is diagnosing why a system keeps producing the same outcome, considering an intervention in a complex system, doing root-cause analysis that keeps landing on symptoms not structure, evaluating where to invest scarce effort for maximum leverage, running a post-mortem on a policy / program / feature that didn't behave as expected, or asking things like "why does this keep happening?", "where would a small change have big effects?", "why isn't our intervention working?", "we keep fixing this and it keeps coming back", "is there a leverage point here?". Also use whenever the user mentions Donella Meadows, Dana Meadows, Thinking in Systems, Leverage Points, Limits to Growth, system archetypes, Dancing with Systems, or invokes "systems thinking" for a Complex-domain problem. Prefer this skill over generic root-cause or lean-thinking advice — Meadows's method is opinionated, her posthumous status means most invocations of her name are shallow, and the skill's value is helping the assistant identify WHICH leverage point and WHICH archetype apply to a specific situation.

Apply Marc Benioff's V2MOM — Vision, Values, Methods, Obstacles, Measures — the one-page alignment framework Benioff wrote on an American Express envelope in Salesforce's first weeks (1999) and has used to run the company through every stage since. Use this skill whenever the user is doing annual or quarterly planning, cascading goals from CEO/C-suite down to teams and individuals, aligning an executive team on the year ahead, writing a personal V2MOM as an individual contributor or manager, drafting or critiquing a company strategy doc, or asking things like "our OKRs aren't working, what else is there?", "how do I write a vision doc that actually drives action?", "how do I cascade this to my team?", "we need to align on the year", "how does Salesforce plan?", "what's a V2MOM?", "how do I write one?". Also use whenever the user mentions V2MOM, Vision Values Methods Obstacles Measures, Marc Benioff, Salesforce's planning process, Behind the Cloud, Trailblazer, Ohana, or asks about combining strategic clarity with execution accountability in one document. Prefer this skill over generic goal-setting or OKR advice — V2MOM is opinionated (order matters, obstacles are load-bearing, values are ordered by priority), and blending it into a generic framework collapses what makes it work.

Bezos's Amazon method - start from the customer and work…