stromy-org/nl-ez-plugin

Render brand-aware data charts (bar, line, waterfall, sankey, sunburst, treemap, funnel, heatmap, gauge, radar and other Plotly types) to PNG/SVG with the client's compiled brand theme, for embedding in PPTX, DOCX, PDF or HTML deliverables. You author the Plotly figure JSON; the server render_chart tool applies palette and fonts and runs a palette check and visual QA. Use for any chart, plot, graph or KPI visual. Not for process flows, org charts or stakeholder maps (use format-diagram), unbranded Mermaid pie/XY in Markdown (use format-mermaid), or a chart native to an Excel workbook (use format-xlsx).

Generate branded diagrams (process flows, architecture views, stakeholder maps, org charts, timelines, funnels, matrices, mind maps, sequence/state/ER/gantt) as SVG/PNG for embedding in PPTX, DOCX, PDF or deck deliverables. Author Mermaid text and call the server-side render_diagram tool (auto-layout, deep brand theming, works in a deployed sandbox, brand-gated). Use when asked to draw, map or visualize a process, structure, workflow or relationship. Not for numeric data charts (use format-chart) or plain Mermaid syntax for a README, wiki or Markdown page (use format-mermaid).

Create, read, or edit Word documents (.docx) — tables of contents, headings, page numbers, headers/footers, letterheads, tracked changes, comments, find-and-replace, and reorganizing or extracting content. Use when asked for a Word doc, a report, memo, letter, brief, or template as a .docx, or to edit or analyze an existing one (branded or not). Client branding applies automatically when an overlay resolves. Not for a net-new branded report, brief or memo where visual quality matters — use format-docx-hd; not for PDFs (use format-pdf), spreadsheets (use format-xlsx), or Google Docs.

High-fidelity branded Word (.docx) creation from authored HTML. Server-renders via the render_docx MCP tool — Chromium parses the HTML, docx-js emits OOXML, and an OOXML layout lint plus a LibreOffice render-back contact sheet run inside the render. No local build script. Uses the invoking plugin's brand overlay. Use when GENERATING a branded Word deliverable: reports, briefs, memos, letters, policy documents, findings write-ups, any client-facing .docx where visual quality matters. Triggers on: 'branded Word doc', 'report in Word', 'branded .docx', 'HD docx', 'client-ready Word document'. NOT for editing an existing .docx, tracked changes, comments or template surgery — use format-docx.

Product guide for producing documents, decks, and other deliverables — what to ask for, which format to pick, how the skills hand off to each other, and where finished work lands. Use whenever someone asks "how do I make a document/deck/PDF", "how do I create a proposal/presentation", "which format should I use", or any orientation question specifically about producing a deliverable. For first-time install, switching on connectors, or a general "how do I get started", the getting-started skill is the entry point. Routes to the right renderer; does not render anything itself.

Author a premium, self-contained branded HTML slide deck: ONE .html you hand-write, with responsive 16:9 stages, scroll-snap navigation, inline-SVG brand motifs and the client's fonts and logos embedded by the server at verification so it opens offline as a single file. Brand-integrated through the client overlay, with seeded variance so decks read as siblings, not clones. Use for an HTML, web or browser slide deck, a shareable single-file presentation, or turning a topic, brief or report into branded slides, even without saying 'HTML'. Not for a reveal.js deck with speaker notes (use format-html-reveal), an editable PowerPoint (use format-pptx-hd) or a paginated PDF (use format-pdf-hd).

Create a premium reveal.js slide deck from a brief or outline as ONE self-contained .html (reveal.js, brand fonts and images all inlined) that opens offline in any browser and shares as a file or link, server-rendered by the render_deck MCP tool with a brand gate and visual QA. Gives a speaker view (S key), keyboard navigation, layout archetypes and a variance engine so every deck stays premium and visibly the client's own. Use for 'reveal.js deck', 'web deck with speaker notes', 'interactive presentation', 'shareable web slides'. Not for a hand-authored single-file HTML deck with full pixel control (use format-html-hd), an editable PowerPoint (use format-pptx-hd), or a paginated print PDF (use format-pdf-hd).

Use when asked to translate a deliverable, add a language version, check whether a translation is current, find out what changed since the last translation, or work out why a copy still has source-language text in it. Maintain a deliverable that exists in several languages without the versions drifting apart: reports which translation units are missing, stale, current, or obsolete; returns only what must be translated; validates a translated batch against protected tokens, placeholders, URLs, figures, and terminology; and hands back a replacement target state to persist. Works on structured translation units a producer supplies, not a raw file. Not for translating a finished Word, PowerPoint or PDF file in place (use that format's skill).

Write Mermaid diagram syntax (flowchart, sequence, class, state, ER, Gantt, mindmap, timeline, C4, Kanban, architecture and other types) as fenced code that Markdown hosts render (README, GitHub/GitLab, wiki, blog), with live preview and PNG/SVG/PDF export when the mermaid MCP is connected. Use when the user wants Mermaid syntax or a quick text-to-diagram sketch. Not for a branded figure inside a PPTX, DOCX, PDF or deck (use format-diagram) and not for numeric data charts (use format-chart).

Use when someone wants animation in a deck or site: an animated reveal or transition, a line that draws itself, a counting KPI, a baked clip from web animation, advice on how much movement a deck should have, or a surface that feels alive without becoming gimmicky. Authors tasteful, brand-aware motion primitives (staggered reveal, line-draw, count-up, motion-path, scroll-reveal, element morph) under a restraint doctrine. Two modes: embed-live (inline a primitive into a deck or site) and bake (turn a self-contained HTML animation into MP4/GIF/PNG via render_motion, for PowerPoint or PDF). Brand-optional (not bake). Not for a narrated or voiced explainer video (format-video-hd), numeric data-viz (format-chart), or structural diagrams (format-diagram).

Comprehensive PDF manipulation toolkit — extract text and tables, create new PDFs, merge or split documents, and fill in forms. Use when a PDF form needs filling, or a PDF must be programmatically processed, generated, split, merged, or analyzed at scale. Not for a new branded PDF deliverable — use format-pdf-hd instead.

Create a high-fidelity branded PDF (proposal, executive brief, brand book, policy report, case study, white paper or one-pager), designed HTML-first with CSS, web fonts, SVG and paged media, and server-rendered by render_pdf behind a brand gate from the plugin's brand overlay. Use when someone asks for a branded, high-quality, HD or magazine-style PDF, or any visually polished client-facing paginated output. Not for filling forms, extracting text, merging, splitting or unbranded PDFs (use format-pdf), nor for structuring a multi-section document with no agreed outline yet (start with format-prepare-document).

DEFAULT PPTX skill for **unbranded deck creation** and **editing / analysis / OOXML / comments / speaker-notes of any existing deck** (branded or not — in-place OOXML edits, not a branded render). For a **net-new branded deck** or render-anchors, use format-pptx-hd (server-rendered + brand-gated). This skill does not render a branded deck from scratch.

Create a new branded PowerPoint (.pptx) deck from scratch from the plugin's brand overlay, server-rendered by render_pptx in the brand fonts behind a brand gate (HTML-first; the deck stays editable in PowerPoint). Use when someone asks for PowerPoint, a .pptx or an editable deck, including pitch, investor or executive decks, and for render-anchors. Not for editing or analysing an existing deck or an unbranded quick deck (use format-pptx; this skill regenerates and cannot edit in place), a browser or offline HTML deck (use format-html-hd or format-html-reveal) or a paginated print PDF (use format-pdf-hd).

Plan a multi-section document before rendering it: section plan, brand-aware structure, complementary charts and diagrams, sign-off, then handoff to the right format-* renderer. Use when someone asks for a proposal, pitch, report, white paper, executive brief or board deck and the structure is not yet decided, or says 'help me plan or outline the document'. Not for a one-page memo, a single slide, an edit of an existing file, or when an approved section plan or envelope already exists: go straight to format-pdf-hd, format-pptx-hd, format-docx-hd or format-html-hd. Branding comes from the invoking plugin's brand overlay.

Author a premium branded explainer VIDEO — a server-rendered MP4 with animated titles, kinetic typography, animated process-flow diagrams, animated data (bars, KPI count-ups), image reveals, captions, and optional voiceover or music. Storyboard-first and accuracy-gated: a scene-by-scene shot list with citations, then a brand-gated render with a per-scene visual review. Brand-required. Use when someone wants a video, explainer, motion-graphics video, or a brief, plan, or report turned into a branded video — even without saying 'video'. Content stays accurate, never fabricated. Not for an HTML-to-MP4 deck/PDF clip (format-motion), generative AI footage (video-production), or an editable branded deck (format-pptx-hd).

Comprehensive spreadsheet creation, editing, and analysis — formulas, formatting, data analysis, and visualization. Use when working with spreadsheets (.xlsx, .xlsm, .csv, .tsv) to create a new one, read or analyze data, modify an existing one while preserving formulas, recalculate formulas, or build a chart. Not for a Word document (format-docx), a PowerPoint deck (format-pptx), or a data-viz chart to embed elsewhere (format-chart).

Day-0 setup and orientation for a Stromy plugin — install it, switch on the connectors it needs, confirm it worked, and find which guide answers your question. Use whenever someone is new, asks how to get started, how to install or set up the plugin, what they have available, or hits one of the two common snags: skills that look out of date after an update, or a tool that reads "Connected" but will not run in the chat. This skill ships inside the plugin and needs no connector, so it is the one that still answers when nothing else is switched on.

Track what Dutch ministers and cabinets formally committed to, and classify each commitment's delivery status (Kept/Compromised/Stalled/Deferred/Broken/Not-yet-testable) with cited evidence, cross-checked against Rijksfinanciën budget appropriations and CBS outcome statistics. Use when the user asks what a minister promised vs delivered, wants a coalition-agreement scorecard, needs toezeggingen delivery tracked for a client brief, or wants a said-vs-done ledger. Do NOT use for adversarial contradiction-hunting or angle-finding on a topic (→ nl-tensions), single-claim fact-checking (→ nl-evidence-grounding) or for open-ended topic/dossier monitoring (→ nl-gov-data). Needs a frozen, bounded commitment corpus — a comprehensive tracker, not a spot-check.

Deep brief on ONE Dutch legislative dossier: current phase, timeline, moties and votes, EU forcing function, and a framed passage forecast (likelihood band, decisive variable, assumptions, two scenarios, base rate). Use for 'dossier tracker', 'will it pass', 'passage signals', 'dossier {number}', 'wet {name}', 'wetsvoorstel', 'TK dossier', 'legislative dossier', 'bill outlook', 'where does bill X stand and will it pass'. Not for: recurring watching or alerts on dossiers (use nl-monitor); individual-MP defection risk or stakeholder maps (use nl-parliamentary-positioning); 6-36 month foresight across topics (use nl-horizon-scan); a quick one-off status lookup with no forecast (use nl-gov-data).

Track how a specific EU directive is being transposed into Dutch law: deadline, national implementing measures (NIMs), BWB article-level correspondence table, gold-plating screen, Wetgevingskalender live pipeline status, and an infringement/Single Market Scoreboard benchmark. Surfaces compliance risk and PA lobbying windows. Use for 'EU directive', 'transposition', 'implementatie richtlijn', 'richtlijn omzetting', 'gold-plating', 'NIM', 'CELEX', 'implementatiewet', 'conformity deficit', 'infringement', 'Single Market Scoreboard', 'EU→NL', 'directive tracker', 'does this directive apply to NL', 'is NL compliant with directive X', 'national implementing measure', 'transposition deadline'.

Fact-check one Dutch policy claim or figure against the official record — CBS statistics, Rijksfinanciën budget phases, kamerstukken, Rechtspraak — into a source-cited Evidence Dossier with a fixed-vocabulary triangulation verdict. Use whenever the user asks 'is this claim true', wants a minister's statement or stated figure verified, wants a number grounded in the official record, or asks what CBS says about X. Handles 1-5 bounded claims — a targeted fact-check, not a monitoring stream. Not for: pledge-delivery tracking across a commitment corpus (→ nl-accountability); contradiction-mining across a topic (→ nl-tensions); a full policy landscape survey (→ nl-policy-legislative-landscape).

Track Dutch public affairs via the nl-gov-data MCP: Tweede Kamer, OpenOverheid, KOOP/CVDR, BWB law, Wetgevingskalender, ROO, Rechtspraak, CBS, Rijksfinancien, EUR-Lex, data.overheid.nl, plus deep-reading and quoting document text. Use for Dutch parliament activity, kamerstukken, dossier numbers, MPs, faction composition, ministry narrative, wetgeving, government organizations, or reading, quoting or summarising a Dutch government document, transcript, law article or attachment; the general-purpose fallback. Not for: single-dossier passage forecast (nl-dossier-tracker), recurring digest or alerts (nl-monitor), stakeholder map (nl-parliamentary-positioning), claim fact-check (nl-evidence-grounding), landscape report (nl-policy-legislative-landscape).

Product guide for researching Dutch government, policy, and legislation — what to ask for, which research skill fits, how findings hand off to your branded documents, and where finished work lands. Use whenever someone asks "how do I research Dutch government/policy/legislation", "how do I turn research into a report/briefing", "which research skill do I need", or any orientation question specifically about Dutch public-affairs research. For first-time install, switching on connectors, or a general "how do I get started", the getting-started skill is the entry point. Routes to the right research skill and hands off to the document tools; does not itself produce documents.

Forward-foresight scan for Dutch policy and regulatory signals over 6-36 months — EU/pipeline/procurement/Senate signals rated on an impact×likelihood grid into a prioritised watchlist. Use when the user asks what is coming in the next 1-3 years, what EU directives will force NL legislation, what signals to monitor for a sector, or wants a horizon scan, foresight brief, or forward radar. Do NOT trigger for: a single known bill's passage forecast (→ nl-dossier-tracker); article-level directive transposition mapping (→ nl-eu-transposition); any bill-status alert, even one-off (→ nl-monitor); a consulting-style report connecting all current activity on one dossier or topic (→ nl-policy-legislative-landscape).

Analyse how a named subject — company, product, organisation, technology, sector, or policy issue — is framed in official Dutch government and parliamentary records via nl-gov-data. Use whenever the user wants official-source reputation or issue framing: how an organisation is portrayed, how a technology is treated in policy, how a contested issue is characterised, what reputational risks surface. Produces narrative analysis, mention/context tables, issue maps, PA/PR implications and DOCX-ready appendices. Not for: strategy or recommendations (out of scope); said-vs-done hunting (→ nl-tensions); single-dossier chronology (→ nl-dossier-tracker); single-claim fact-checking (→ nl-evidence-grounding); pledge tracking (→ nl-accountability).

Recurring monitoring and alert product for Dutch parliamentary and legislative activity. Watches a defined topic/dossier/actor matrix over weekly windows and produces a weekly briefing: breaking alerts, dossier-status, forward agenda. Use whenever the user wants a topic or dossier set monitored on a cadence, asks for a weekly brief, asks 'what changed', wants a threshold alert on a watched dossier, or wants a re-runnable digest. Not for: single-dossier passage forecasting (→ nl-dossier-tracker); 6-36 month foresight (→ nl-horizon-scan); toezeggingen delivery tracking (→ nl-accountability); contradiction hunting (→ nl-tensions); a one-off pull-up-and-summarise on a named dossier with no recurring cadence (→ nl-gov-data).

Map Dutch parliamentary positioning and official political posture from nl-gov-data. Use whenever the user asks which MPs, factions, committees, ministries, votes, motions, debates or official bodies are active around a Dutch policy topic. Mode A produces PA/PR stakeholder maps, activity summaries and DOCX-ready client analysis. Mode B produces a per-legislator meeting-prep disposition brief — trigger on 'MP profile', 'disposition brief' or 'meeting prep for <MP>'. This is official institutional positioning, not public sentiment. Not for: entity or issue framing (→ nl-issue-framing); a said-vs-done contradiction on one actor (→ nl-tensions); a single-dossier deep-dive (→ nl-dossier-tracker); a bounded pledge ledger (→ nl-accountability).

Use whenever the user wants the big picture on a Dutch public-policy topic: a policy landscape, legislative scan, dossier map, official activity timeline, regulatory or PA/PR baseline, or a consulting-style report. Builds official-source analyses from the nl-gov-data MCP, connecting Tweede Kamer, KOOP, BWB, Wetgevingskalender, ministries and committees into a client-ready landscape. Do NOT use for a single named entity or product (→ nl-issue-framing), who moves a decision (→ nl-parliamentary-positioning), said-vs-done or a contradiction (→ nl-tensions, nl-accountability), one figure/claim (→ nl-evidence-grounding), one dossier's passage (→ nl-dossier-tracker), 6-36 month foresight (→ nl-horizon-scan) or recurring alerts (→ nl-monitor).

Detect contradictions, said-vs-did gaps and non-obvious tensions in Dutch public policy via the nl-gov-data MCP. Produces structured, source-cited tension reports — dormant dossiers, expired motions, coalition flips, commitments not kept, claim-vs-structure mismatches. Use whenever the user asks about tensions, contradictions, gaps, what's not adding up, said-vs-spent, where a policy is stuck, or angles for a roundtable on a Dutch policy topic. Not for: monitoring (→ nl-gov-data); strategy (out of scope); toezeggingen or commitment delivery tracking across a defined corpus (→ nl-accountability); single-claim fact-checking (→ nl-evidence-grounding); entity or issue framing (→ nl-issue-framing).