Menu
Skills in this repository
SkillsMP has collected 228 skills from hpsgd/turtlestack. Open a skill to review its source and details.
hpsgd/turtlestackShowing 40 of 228 collected skills.
Write a one-page product vision using Roman Pichler's Product Vision Board (vision, target group, needs, product, business goals). Use when starting a product, entering a new market, or when a team has lost strategic coherence and needs alignment on direction…
Audit delivery status for watermelon reporting — green outside, red inside. Cross-checks status reports against the RAID log against the actual delivery state, and surfaces the cultural conditions that drive watermelon reporting (red triggers punishment…
Coordinate the upstream pieces a release needs before the release-manager's gates — support briefed, GTM aligned, ops runbook drafted, governance approvals lined up, customer comms planned. Assembles a release-readiness package and hands it to the…
Facilitate a scrum of scrums — a 15-30 minute cross-team coordination meeting with one representative per team covering what each team did, what is next, and cross-team blockers. Includes the theatre-vs-real check (representatives who cannot speak for their…
Forecast a delivery using reference-class forecasting — historical actuals-vs-estimates from similar past deliveries rather than bottoms-up team estimates under pressure. Corrects the planning fallacy (Kahneman/Lovallo, via Flyvbjerg). Use when a date or…
Prepare for a GDS service assessment — the alpha/beta/live phase gate. Compiles evidence against the GOV.UK Service Standard, runs a mock assessment, and tracks remediation of gaps. Use when a digital service approaches a phase gate (end of alpha, end of…
Prepare a steering-committee pack — a fortnightly or monthly governance pack pitched at the right level of abstraction, with decision asks named explicitly. Not a status read-out. Use for steering-committee meetings, executive updates, or any governance forum…
Run the weekly RAID log review — scan for status changes, close resolved items, escalate overdue ones, and run the item-rot, risk-vs-outcome, RAG-kabuki, and assumption-paralysis diagnostics. Use for the weekly delivery review or whenever the RAID log needs a…
Create or update the live dependency map — what each team needs from other teams, with owning team, contact, status (on track / at risk / blocked), and resolution date. Produces team-level and programme-level views and flags escalation triggers. Use for…
Write or add to a RAID log — risks, assumptions, issues, and dependencies for a delivery. Structures each item correctly (risk as cause not outcome, assumption with validation owner, issue with 48hr escalation, dependency with contact and date) and enforces a…
Write a weekly delivery status report with four components — what happened, what is at risk, what decisions are needed, what help is asked for. Assigns honest RAG (red = won't hit target without intervention) and attaches a road-to-green action plan when…
Assess a team against the Team Topologies four team types (stream-aligned, enabling, platform, complicated-subsystem) and three interaction modes (collaboration, X-as-a-service, facilitating). Produces a topology read and recommendations on team type and…
Diagnose deep agile dysfunction using Aino Corry's retrospective anti-pattern catalogue, the Zombie Scrum four-symptom check, and Ron Jeffries' Dark Scrum diagnostic. Distinguishes team-fixable problems from organisational ones. Use when ceremonies look fine…
Observe a full sprint cycle and name ceremony anti-patterns with evidence — status-theatre standups, secretary Scrum Master, planning-as-task-assignment, retrospectives without action items. Produces an audit with prioritised coaching actions. Use when a…
Help a multi-team setup pick or deliberately reject a scaling framework — where LeSS, Nexus, Scrum@Scale, and SAFe fit and where they don't. Produces a recommendation grounded in the coordination problem, not framework fashion. Use when leadership is…
Facilitate the team to author its own Definition of Done — specific, testable, collectively owned, not a management wish list. Runs a specificity test and ties the DoD to the release-manager's gates. Use when a team has no DoD, a vague one, or is presenting…
Teach the team to read its own flow data — cycle-time scatterplot, throughput, work-item age, and cumulative flow diagram — with Little's Law as the why behind WIP limits. Produces a flow read and coaching plan. Use when a team is drowning in WIP, has…
Coach a team on the Kanban Method's six core practices (visualise, limit WIP, manage flow, make policies explicit, feedback loops, evolve experimentally), when Kanban beats Scrum (discovery, ops, irregular arrival), and the Scrumban hybrid. Use when a team's…
Facilitate the team to author its own working agreements — team-generated, specific, and testable — with a review cadence. Distinguishes working agreements from the Definition of Done. Produces a working-agreements artifact. Use when a team has no agreements,…
Design and facilitate a team retrospective using Derby & Larsen's five phases, with a format catalogue (Start/Stop/Continue, 4Ls, Sailboat, Mad/Sad/Glad, 5 Whys, Lean Coffee) and Liberating Structures. Produces a retro plan and an output file with action…
Facilitate Sprint Planning across its three topics (why the sprint is valuable, what can be done, how it gets done). The team authors the Sprint Goal; the coach facilitates, never writes it. Names planning anti-patterns. Use when running or coaching a sprint…
Coach the sprint review as a collaborative working session that inspects the increment and adapts the backlog — not a demo to management. The product manager runs the review; the coach facilitates the working-session craft. Use when a sprint review has become…
Run a team-health scan: an Edmondson-style psychological-safety questionnaire, a working-norms review, and a Tuckman-stage assessment on reteaming. Produces a health report distinguishing what to act on from what to surface. Use to establish a safety…
Mine account-level voice-of-customer themes from churn, at-risk, QBR, renewal, expansion, and cancellation evidence. Use to extract the qualitative retention narrative behind churn and feed retention themes to the roadmap. The customer-success VoC lens —…
Prepare for an industry-analyst briefing (Gartner, Forrester, IDC) — briefing deck structure, proof points mapped to evaluation criteria, demo flow, and Q&A prep for hard analyst questions. Use before a vendor briefing, Magic Quadrant / Forrester Wave…
Build structured pricing intelligence across the competitive set — list prices, packaging, tiering, add-ons, discounting signals, and the value metric each competitor charges on. Use when you need a clear picture of how the market prices and packages before a…
Design a willingness-to-pay or pricing research study — selects Van Westendorp, Gabor-Granger, conjoint, or MaxDiff based on the pricing question being asked, then produces the survey design (sample, questions, analysis plan). Use when you need evidence to…
Capture the go-to-market voice-of-customer lens — how the market reacts to positioning, what objections sales hears, and why customers switch to or away from us. Use when you need market-facing VoC (positioning resonance, objection patterns, switching…
Build a messaging hierarchy — the primary message, supporting messages, proof points, and per-persona variants — that flows from an existing positioning. Use when you have positioning and need the message architecture that turns it into consistent copy across…
Write a strategic narrative using the Andy Raskin structure — name the big change in the world, name the stakes of the old world (the enemy), paint the promised land, and prove you can get the buyer there. Use for keynote, pitch, vision deck, or…
Analyse retention curves and cohort behaviour, cutting by segment and benchmarking the curve shape against what is healthy for the product type. Use when measuring retention, comparing signup cohorts, diagnosing whether a product has product-market fit, or…
Define a product's North Star Metric and its supporting input metrics, with explicit checks against Goodhart's Law and multi-customer-type incoherence. Use when establishing the single metric that captures delivered customer value, or auditing an existing…
Design an A/B test or holdout experiment to establish causation, including hypothesis, sample-size calculation, and the evidence hierarchy that weights observed behaviour over stated intent over opinion. Use when designing an A/B test, sizing an experiment,…
Build a product metric hierarchy using HEART or AARRR depending on product type, decomposing the North Star into actionable input metrics tied to OKRs. Use when structuring product metrics into a tree, choosing between engagement and funnel frameworks, or…
Specify the analytics events, properties, identity model, and attribution needed to measure a product's metrics, as a hand-off to the data-engineer for implementation. Use when defining what events to track, designing an event schema, or preparing a tracking…
Map the assumptions a bet depends on on two axes — how much we know (knowledge) and how much it matters if we're wrong (impact). Produces a prioritised list of which assumptions to test first. Use before committing engineering effort to a solution, to find…
Define an Ideal Customer Profile from firmographic and behavioural criteria, built from current best customers and lost/churned accounts. Produces an account-level ICP distinct from an individual-level persona. Use to focus discovery, qualification, and…
Design a pretotype experiment from Alberto Savoia's catalogue (Mechanical Turk, fake door, smoke test, concierge MVP, Wizard of Oz) plus a Market Engagement Hypothesis in XYZ format. Produces a runnable experiment that generates behavioural data before…
Validate discovery hypotheses against Voice-of-Customer signal across tickets, reviews, churn reasons, surveys, and sales feedback. Produces a synthesis that confirms, contradicts, or qualifies a hypothesis with triangulated VoC evidence. Use to pressure-test…
Run a Bob Moesta JTBD switch interview — reconstruct the timeline of a real purchase and code the four forces (push, pull, anxiety, habit). Produces a force-coded timeline of why a specific customer actually switched. Use to understand real buying decisions…