Claude Skills by gethamster

OKR mistakes to avoid: spot common OKR anti-patterns such as sandbagging, tasks as key results, too many OKRs and pay-linked scores, and fix each one.

Keep autonomous squads pointed the same way: make priorities visible, write a few enabling constraints, and review alignment on a regular cadence.

Balance helpfulness and harmlessness in Constitutional AI: cut over-refusal and preachy answers while keeping the model safe.

Benchmark AI product pricing against competitors: normalize prices to one value unit, compare what each price buys, and choose a position.

Break a GIST step-project into small, sprint-ready tasks that each trace back to the experiment, the idea and the goal they serve.

Build a reusable 4Ls retrospective board and template, physical or digital, with prompts, voting space and an action area the team reuses.

Build a customer journey inventory: catalog every journey your teams know about into one journey portfolio with owners, levels, evidence, and status.

Plan the RACE framework Reach stage: choose awareness stage marketing channels across paid, owned and earned media, and check the right audience arrives.

Plan a content waterfall strategy: one pillar asset cut into derivative pieces and released across channels in a set order, then refined by response.

Start and sustain voluntary, company-wide guilds that share knowledge, tools and decisions across tribes, using what research on Spotify's guilds found.

Create a customer journey map of the five stages from real research, run the mapping session, and keep the map in use for planning.

Build a GO product roadmap template, Roman Pichler's goal-oriented roadmap template, with date, name, goal, features, and metrics rows.

Build a HEART metrics dashboard that shows a short, prioritized set of UX metrics by category, with baselines and release notes teams review.

Build a promotion strategy in the marketing mix: an integrated promotion plan that coordinates advertising, content, PR and sales around one message.

FDE skill development: assess and train the engineering, data, domain and communication skills a forward deployed engineer needs on site.

How to build a minimum viable product: scope the smallest version that tests your riskiest assumption, instrument it, launch it and decide what's next.

Build a North Star Metric dashboard that puts the metric beside its input metrics, shows trends with context, and drives a weekly review with owners.

Tracking a North Star metric on a dashboard shaped like the framework: business results, the metric, inputs, system health and bets on one page.

Build a Now Next Later roadmap template with three horizon columns, problem-framed cards and objective tags that a team can fill in and keep current.

Build an outcome-based roadmap presentation that shows stakeholders the objectives, outcomes and bets behind planned work, and wins their buy-in.

Build a planned journey funnel diagram that shows stakeholders how buyers flow through each stage, where drop-off happens and why.

Build a PM career development plan from competency gaps on Cabage's grid, using a PM growth plan template built on focus areas, real work and reviews.

Build post-purchase loyalty loops: turn passive loyalists into active ones who repurchase without re-evaluating and recommend you to others.

Build a RACE planning template: a digital marketing plan with objectives, tactics, owners, budgets and KPIs for Plan, Reach, Act, Convert and Engage.

Build a sprint prototype in one day: split the storyboard across Makers, a Stitcher, a Writer and an Asset Collector, then trial-run the facade.

Building a roadmap from MoSCoW outputs: turn project, increment and timebox priorities into MoSCoW phased delivery with releases and review points.

Build a reusable retrospective template that maps activities, timings and facilitator cues to the five phases, so anyone can run a solid sprint retro.

Build a start stop continue retrospective template for a whiteboard, doc or worksheet, with prompts, voting space, an action table and a free text box.

Build business systems in the E-Myth Manager role: document, test and hand off repeatable business processes so work runs the same way without you.

Build Claude topic clusters where every page answers a distinct reader need, checked against search data and free of near-duplicate pages.

Build topical authority that LLMs recognize: cover a few subjects in depth, link the pages clearly and earn independent coverage that confirms it.

Build topical authority with a pillar and cluster content strategy: choose one topic, cover it completely, link it tightly, and grow it from evidence.

Design one reusable V2MOM template with built-in prompts, ranking, parent links and owner fields, so documents at every level compare cleanly.

Calculate AI inference unit economics: the fully loaded cost per request and per unit of customer value that every AI price and tier rests on.

Calculating RICE scores turns reach, impact, confidence and effort estimates into one ranked backlog, then checks that ranking before a team commits.

Calibrating confidence ties each RICE Confidence value to the evidence behind the reach, impact and effort estimates, so hope cannot outrank data.

Turn the company V2MOM into linked department, team and individual V2MOMs, with every method and measure traceable to the level above.

Categorize and prioritize retrospective feedback: cluster Start, Stop and Continue items into themes, dot vote, and cut the list to a few owned actions.

Sort team feedback into the 4Ls categories, tell Liked, Learned, Lacked and Longed For apart, and resolve notes that fit more than one.

Now Next Later categorization: turn a backlog into problem-framed initiatives and place each in Now, Next or Later by confidence and priority.

How to categorize requirements into the MoSCoW categories with the DSDM tests, including where the Must have vs Should have line falls.

Choose between AI pricing models (seat, usage, credits, outcome or hybrid) by scoring each against your cost profile, value unit and buyer.

Compare Double Diamond vs design thinking and Lean UX, then choose the framework, or the combination, that fits your project and team.

Choose between Scrum, Kanban and Scrumban by reading how your team's work arrives, then confirm the choice with a short, measured trial.

Choosing kanban software: evaluate kanban board tools against the practices you run, test finalists on real work, and pick one you will not outgrow soon.

Choosing retrospective activities: pick one exercise per phase to suit the sprint, the team and the time, so each retro surfaces something new.

How to end a retrospective: read back the action items and owners, thank people, collect a quick ROTI on the session and close on time.

Agile coaching for teams in adoption: map how they work today, add one practice at a time with their consent, and hand the process over to them.

Collect six forces model data with a market research process that sources, checks and files evidence for every force before anyone rates it.

Communicating roadmaps to stakeholders: present a Now Next Later roadmap to executives, sales, teams and customers without implying dates.