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.
Pattern-code a window of discovery interviews into themes, update opportunity solution tree nodes, and flag theme saturation. Produces a synthesis that turns raw interview notes into discovery decisions. Use after running several interviews to convert…
Set up a Teresa Torres continuous-discovery cadence — weekly customer interviews with the product trio, automated recruiting, and calendar discipline. Produces a discovery plan that runs continuously rather than as a one-off research sprint. Use when standing…
Write a generative discovery interview guide following The Mom Test — questions about specific past behaviour, not hypothetical future intent. Produces a runnable guide with opening, behavioural questions, fluff-redirect prompts, and a close. Use before…
Define a Jobs-to-be-Done analysis for a product or feature area. Produces structured job statements, outcome expectations, and hiring/firing criteria. Use when entering a new problem space, validating product-market fit, or reframing features as customer jobs.
Build or update a Teresa Torres Opportunity Solution Tree — desired outcome at the root, opportunity space from real interviews, solutions per opportunity, and assumption tests. Produces a visual artifact that makes discovery decisions explicit. Use when…
Write a Product Requirements Document from rough notes or a feature idea. Produces a structured PRD with problem validation, user definition, RICE prioritisation, acceptance criteria, and scope.
Build an outcome-shaped product roadmap using Now/Next/Later or GIST. Produces a roadmap that promises changes in customer behaviour, not a feature timeline with dates. Use when communicating product direction, planning a quarter, or replacing a date-based…
Run a backlog refinement event — facilitate the team through INVEST checks, a capacity budget, dependency surfacing, and a Definition of Ready gate so items leave the session sprint-ready. Use to prepare and run a live refinement ceremony, not a solo backlog…
Extract Voice-of-Customer product themes from support tickets — code ticket text into product themes, separate UX-failure signal from doc-gap signal, weight themes by frequency × severity, and surface a ranked theme list for the roadmap. Use when…
Design an ongoing voice-of-customer (VoC) programme combining NPS, CSAT, CES and qualitative synthesis. Produces a programme blueprint covering metric selection, survey design (wording, sampling, cadence), closing-the-loop routing, and a method for tying…
Deep-dive analysis of a SINGLE competitor (15-30 pages) triggered by a strategic event — a raise, a launch, entry into your segment, or a pricing change. Covers product architecture from public sources, pricing, ICP, GTM motion, funding, and key people.…
Run Porter's Five Forces (supplier power, buyer power, new entrants, substitutes, rivalry) to judge whether an industry or segment is structurally attractive to enter or stay in — and where the real pressure sits. Writes a conforming report (per…
Run a PESTLE macro-environmental scan (political, economic, social, technological, legal, environmental) for a market or expansion decision — then argue which forces actually matter and what they imply. Writes a conforming report (per report-conventions) to…
Mine G2, Capterra, app stores, and Trustpilot reviews for OWN and COMPETITOR products: extract themes, quantify their weight, and segment-skew the findings. Writes a conforming report (per report-conventions) to <engagement_dir>/review-mining/<slug>.md. Use…
Build a Blue Ocean strategy canvas (Kim & Mauborgne): plot your value curve against competitors across the factors the industry competes on, then drive an ERRC grid (Eliminate / Reduce / Raise / Create) off it. Writes a conforming report (per…
Build a Wardley Map (Simon Wardley): plot the components of a value chain on visibility (user need at top) x evolution (genesis -> commodity) axes to expose where commoditisation is approaching, where competitors over/under-invest, and where build/buy/partner…
Analyse why deals were won or lost from the BUYER's perspective using third-party-style retrospective interviews. Surfaces the gap between sales-reported and buyer-reported loss reasons. Writes a conforming report (per report-conventions) to…
Produce a periodic (quarterly/annual) executive-audience market landscape report that stitches existing market-sizing and competitive-analysis outputs into one narrative: market definition and size, competitive set, recent moves, and a trend scan. Writes a…
Render a meeting Q-and-A document as a printable PDF for note-taking on a Remarkable Paper Pro. Cover page is a fillable form (date, time, attendees, apologies, topic allocations); subsequent pages give each agenda item half a page with a NOTES area and a…
Upload a PDF to a reMarkable device (Paper Pro, rM2) wirelessly via the reMarkable cloud. Wraps the ddvk/rmapi CLI inside a Docker image — Docker is the only host requirement. Requires a one-time interactive pairing (--pair) before normal uploads work. Falls…
Render a markdown document as a brand-styled A4 PDF (hps.gd typography and palette). Optional cover page from YAML frontmatter (title, subtitle, metadata). Use --style <name> for a bundled stylesheet (default: report) or --css <path> for a project-specific…
Retrieve content from a URL using a four-tier escalation strategy: WebFetch → curl → Playwright (JS rendering) → human escalation (anti-bot bypass). Use when standard WebFetch fails or when the target URL is known to require JS rendering or bot mitigation…
Save a web page as a rendered PDF using shot-scraper (Playwright/Chromium). Use to archive primary-source evidence for research — bot-protected and JS-rendered pages included. Different job from /web-tools:content-retrieval: this produces a PDF file for the…
Research a topic, question, or domain using web sources. Three tiers: quick (5 min answer), standard (structured overview), deep (exhaustive with source cross-referencing). Use when you need to understand an unfamiliar topic before deeper analysis.
Write an event-sourced aggregate using EventSauce — aggregate class, domain events, and state replay.
Write a Behat feature specification in Gherkin with step definitions.
Write a symfony/messenger command, query, or event handler with constructor injection and bus dispatch.
Review PHP code against team conventions — type safety, immutability, testing, linting, error handling, and code structure. Auto-invoked when reviewing .php files.
Scan an engagement directory for conforming research reports, group them by category, and produce a single DOSSIER.md plus brand-styled PDF that embeds every included report. Each embedded report's title becomes an h1 page break, so every report starts on a…
Generate the executive summary section for a dossier. Interactive: the user picks which input reports feed in, names the audience and posture, signs off on an outline, then reviews the draft before it's written into DOSSIER.md. Call standalone to regenerate…
Review Go code against team conventions — error handling patterns, interface design, goroutine safety, and table-driven tests. Auto-invoked when reviewing .go files.
Create a new agent plugin following the standard template. Handles directory structure, plugin.json, agent definition, marketplace.json, README, and RATSI updates.
Search public records for a named individual: court records, business registrations, property (where public), professional licences, and electoral roll. Writes a conforming report (per report-conventions) to…
Investigate a domain's registration, DNS, certificates, hosting, and history using passive public sources. Writes a conforming report (per report-conventions.md) to <engagement_dir>/domain-intel/<domain-slug>.md. Use when mapping a domain's infrastructure or…
Map an organisation's complete public digital presence: domains, web presence, social profiles, apps, code repositories, job postings, and regulatory filings. Writes a conforming report (per report-conventions) to…
Verify that a named individual is who they claim to be, or resolve ambiguity between people sharing a name. Anchors on verifiable claims and cross-references independently. Writes a conforming report (per report-conventions) to…
Investigate an IP address: ownership, hosting provider, ASN, reputation, and associated infrastructure. Writes a conforming report (per report-conventions) to <engagement_dir>/ip-intel/<ip-slug>.md. Passive sources only — no active scanning.
Map publicly visible social media presence for a person or organisation across platforms. Public content only — no access to locked, private, or friends-gated content. Writes a conforming report (per report-conventions) to…
Map a company's ownership chain, related entities, and director networks. Writes a conforming report (per report-conventions) to <engagement_dir>/corporate-ownership/<entity-slug>.md. Use for beneficial ownership investigation, M&A research, or understanding…
Structured overview of a named individual's public presence: professional history, news coverage, academic work, people search aggregators, and company affiliations. Writes a conforming report (per report-conventions) to…