ai-plugins

How to build Pigment Frames, custom full-page JavaScript visualisations that read Pigment data through the PigmentSDK inside a sandboxed iframe. Use when creating or editing a Frame.

Always use when creating or editing a Board. This skill includes supporting files in this directory - explore as needed.

Always use this skill when creating or editing Views, or needing to pick a View.

Use this skill when integrating external data into Pigment - importing an attached CSV file, deciding whether to import into dimensions vs transaction lists, mapping source columns to properties, configuring cross-application (P2P) imports, or troubleshooting data imports. For the step-by-step CSV file import, read data_import_csv.md. Do NOT use this skill for formula updates or list creation unrelated to a data import. This skill includes supporting files in this directory - explore as needed.

Always use this skill when designing, applying, or debugging Access Rights and security in Pigment applications. Provides the AR mental model (User, Role, dimension axis, AR Metric, Apply rule), the canonical decision order, mandatory formula patterns (IFDEFINED guard, BLANK over FALSE), multi-app AR, debugging "why can this user see / not see this data?", and security governance. AR is part of model architecture, not an afterthought.

Always use this skill when writing, editing, or debugging Pigment formulas — including conditional logic, blank handling, date-range logic, aggregation, prior-period lookups, and dimensional transformations. Pigment uses a proprietary formula language — NEVER assume you know the syntax, and ALWAYS read the documentation before writing any formula. Covers data types, modifiers, functions, calculation patterns, and performance trade-offs (calibrated by formula complexity). This skill includes supporting files in this directory; explore as needed.

Best practices for metric default formatting. Covers: decimals, prefix, suffix, currency ($/€), percent (%), K/M/bp/thousand/million scaling, thousand separator, sign / zero / negative handling, text mode (Text / Rich Text / URL / Image / LocaleDateTime), boolean display (checkbox / button). Load when creating or updating a metric's default format. Use cases: format as, display as, show as percent, in millions, two decimals, no decimals, prefix with $, add currency, as K / M / bp, rich text, checkbox, ratio, url, multiplier.

Always use this skill when designing or modifying Pigment applications. Provides the mental model of a Pigment app (Application, Dimensions, Calendars, Metrics, Transaction Lists, Tables), the core concepts (dimension list vs property vs transaction list, metric vs table, sparsity, scope), the canonical order of architecture decisions, a minimal viable application pattern and pointers to deeper-dive docs (architecture, naming, hierarchies, calendars, principles, folders, subsets, performance).

Always use this skill when troubleshooting slow calculations or timeouts, analyzing profiler output to identify bottlenecks, understanding scope propagation, managing sparsity, optimizing formula performance, improving iterative calculations, optimizing access rights performance, conducting systematic performance audits, auditing a Pigment application (modeling, formula hygiene, folders, boards, governance), cleaning unused dimensions, metrics, tables, properties, or boards, identifying dead or stale boards, or removing unused metrics. Modeler-agent skill for Performance Insights tools (performance_profile_change, get_top_blocks_by_performance), then classify and fix. Provides the optimization loop, audit vs cleaning modes, and routing to deep dives. Always profile before formula changes; never optimize from assumptions.

Always use this skill when the user mentions or implies versions, Actual, Actuals, Budget, Budgeting, Forecast, Reforecast, Rolling Forecast, Version, Versioning, Plan, switchover, scenarios, snapshots, planning cycles, Actual/Plan layering, plan vs actual, "create version dimension", "set up versioning", or asks for Actual Budget Forecast best practices — or when they extend realized data into a plan or budget (Actual/Budget/Plan layering, forward forecast from actuals) or need to combine or compare actual and plan versions and periods. Covers Version Dimensions (foundational to all planning applications), Native Scenarios (what-if), and Snapshots (freeze data).

Always use this skill when building or extending models for specific planning domains — FP&A, Workforce Planning, Sales Performance Management, Supply Chain Planning, or Financial Consolidation. Covers proven modeling patterns and domain-specific best practices. This skill includes supporting files in this directory - explore as needed.

Always use this skill when querying, exploring, or analyzing existing data in a Pigment workspace. Covers the analysis workflow, query formulation, data concepts, analysis patterns, ambiguity handling, and result interpretation.