OfficeIMO Document Tools

OfficeIMO — Office and document libraries for .NET

OfficeIMO is a family of COM-free .NET libraries for creating, reading, editing, converting, and exporting Office and document formats. It runs in services, desktop applications, build agents, containers, and automation hosts without Microsoft Office, Excel, PowerPoint, Visio, or LibreOffice automation.

This is not one facade over a collection of unrelated document libraries. OfficeIMO owns its OneNote, PDF, Markdown, RTF, OpenDocument, AsciiDoc, LaTeX, OPML, DocBook, CSV, EPUB, ZIP, drawing, Apple iWork source, legacy Word .doc, legacy Excel .xls, and legacy PowerPoint .ppt/.pot/.pps implementations. Word, Excel, and PowerPoint use the Open XML SDK for package mechanics; HTML exposes owned document and provider contracts while retaining AngleSharp for parsing and CSS execution. Converters compose the same first-party object models used by the native packages and return diagnostics when a target format cannot carry everything from the source.

Applications should keep OfficeIMO packages on the same coordinated version. Converters compose package-owned document models and expose result-bearing APIs when callers need fidelity diagnostics.

For existing Windows x64 applications that import through ODBC, OfficeIMO.Odbc exposes bounded .xlsx workbook snapshots as read-only SQL tables. It is built and deployed separately from the .NET document packages.

Upgrading an existing application? The OfficeIMO migration guide covers package, API, and behavior changes across every format. Release history and downloadable artifacts are published through GitHub Releases.

If OfficeIMO saves you time, please consider supporting the work through GitHub Sponsors or PayPal. PowerShell users should start with PSWriteOffice.

Sponsors

See all sponsors

Dependency model

OfficeIMO keeps document engines first-party and optional integrations isolated. The table calls out direct non-OfficeIMO runtime dependencies that matter to package selection; Microsoft/BCL compatibility packages are still used where older target frameworks need platform APIs.

Package family	Direct external runtime dependency	What OfficeIMO owns
Drawing, OneNote, Markdown, RTF, OpenDocument, AsciiDoc, LaTeX, OPML, DocBook, CSV, EPUB, ZIP	No third-party document engine	Parsing, object models, writing, rendering primitives, safety limits, and diagnostics
CAB and passive MSI extraction	None	Managed Cabinet decoding, embedded MSI payload mapping, bounded extraction, and path validation
JavaScript and browser assets	None	Strict TypeScript document layers, streaming XLSX/CSV and embedded portable scripts
Apple iWork source reading	No third-party document engine	Bounded package, Snappy/IWA, protobuf-envelope, record-preservation, and Pages/Numbers/Keynote projection layers
Word, Excel, PowerPoint	Open XML SDK	Fluent/editable object models, lifecycle, validation, conversions, managed image export, and first-party .doc/.xls/.ppt support
HTML document contracts (OfficeIMO.Html.Core)	None	Owned nodes, immutable snapshots, edits and parser/charset contracts
HTML syntax provider (OfficeIMO.Html.AngleSharp)	AngleSharp and System.Text.Encoding.CodePages	Translation to owned nodes, selector/serialization services and charset-provider integration
HTML conversion and rendering	AngleSharp and AngleSharp.Css	Resource policy, media filtering, layout scene, and PNG/JPEG/TIFF/SVG/WebP output; opt-in bridges add RTF, MHTML, email-image, and PDF workflows
PDF	No third-party PDF or cryptographic dependency	PDF parsing/writing/rendering, password security, signature structure, preservation policy, limits, and diagnostics
Email, email stores, and address books	System.Text.Encoding.CodePages	EML/MIME, MSG/OFT, TNEF, mbox, PST/OST, OLM, EMLX, Outlook OAB, MAPI projection, protected-wrapper preservation, limits, and diagnostics
Optional Security provider	Bouncy Castle and System.Security.Cryptography.Xml	CMS/S/MIME/RFC 3161/X.509/XML DSig orchestration behind one typed provider explicitly supplied to Word, PDF, or Email
Optional ChartForgeX bridge	ChartForgeX	Vector-first visual artifact conversion, OfficeDrawing fidelity reports, and Word, Excel, PowerPoint, and PDF placement
Visio	System.IO.Packaging	VSDX/VSTX/VSSX and macro-enabled package model, bounded legacy binary import, diagram builders, editing, validation, topology, and PNG/JPEG/TIFF/SVG/WebP export
Reader.Yaml	YamlDotNet	Reader projection, chunking, limits, locations, and diagnostics
MarkdownRenderer.Wpf	Microsoft WebView2	Rendering shell, presets, plug-in model, and WPF host contract
Optional document AI	OfficeIMO.AI and IntelligenceX adapter	Read-only questions, summaries, field extraction and proposed structure with bounded source evidence; .NET 10 and caller-selected model execution
Optional OCR providers	A caller-supplied executable or an installed Tesseract CLI	Neutral request/result/geometry contracts, bounded provider execution, and opt-in Reader/PDF integrations
Google Workspace packages	System.Text.Json and platform HTTP/cryptography	Credentials abstraction, request/retry logic, Drive placement, translation plans, and reports; no Google client SDK
Converter packages not listed above	Only the OfficeIMO format packages they connect	Feature mapping, limits, loss reports, and destination APIs

At a glance

Surface	Current repository coverage
Modern Office authoring/editing	.docx, .xlsx, .pptx, .vsdx, .vstx, .vssx, .vsdm, .vstm, .vssm
First-party legacy binary support	Word 97–2003 .doc, Excel BIFF8 .xls, PowerPoint 97–2003 .ppt/.pot/.pps
First-party Apple iWork support	Modern IWA-based .pages, .numbers, and .key read/inspect plus editable or visual-fallback projection; bounded template-free Keynote text/image/table slides, Pages body documents and Numbers scalar-table spreadsheets
First-party offline OneNote support	Desktop/FSSHTTP .one, .onetoc2, .onepkg
Managed PNG/JPEG/TIFF/WebP/SVG document export	Drawing; Word, Excel, PowerPoint, HTML, OneNote, Visio, and PDF; HTML-backed email and EPUB; ODT/ODS/ODP through their Office adapters

The checkboxes describe the exact level of support: authoring, editing, reading, preserving, inspecting, converting, or exporting. A checked inspection or preservation item is not presented as full authoring support.

OfficeIMO Studio

OfficeIMO Studio puts the OfficeIMO engines in an app: read, search, mark up and fill PDFs, organize pages, make scans searchable, convert Office documents to PDF, and protect and sign the result. Documents are processed on your own device; connecting the document assistant to a model is optional.

• Windows and Linux: the desktop workspace, from the Microsoft Store or as a signed MSI, Debian package or portable ZIP for x64 and Arm64 from GitHub Releases. It adds comment threads, redaction, content editing, form authoring, conversion from PDF, comparison and a document assistant. See the Studio README.
• Mac, iPad and iPhone: native apps on the same engine, in development, with system document handling, Apple Pencil markup, the document camera, on-device text recognition with Vision, and Shortcuts. See the Apple app README.

The feature table on the Studio page shows what Studio does on each system and what is planned.

The OfficeIMO command-line tool runs the same conversions, extraction and workflows from scripts, and serves them to AI agents over MCP.

OfficeIMO for AI agents

Claude Code, Codex, VS Code with GitHub Copilot, Cursor and other MCP clients use OfficeIMO to inspect, search, fetch and convert local Word, Excel, PowerPoint, PDF and mail files, and to make separate PDF outputs. The OfficeIMO Document Tools plugin adds two Agent Skills and the MCP server, which runs on your computer and opens only the folders you allow:

claude plugin marketplace add EvotecIT/OfficeIMO
codex plugin marketplace add EvotecIT/OfficeIMO --ref master --sparse .agents/plugins

The AI agents page has one-click installation for VS Code and Cursor, the configuration for other clients, every tool the server provides, and the directories that list OfficeIMO.

Packages and tools

Every checked item below is implemented today. Detailed behavior, examples, and fidelity boundaries live in each project README.

Native formats and shared foundations

OfficeIMO.Core

• Common Save, SaveAsync, SaveCopy, ToBytes, and ToStream lifecycle contracts used across formats
• Immutable RGBA colors, named colors, hexadecimal parsing, palettes, and cross-format visual themes
• Image identification, dimensions, MIME metadata, fit modes, projection, cropping, and transform helpers
• Bounded async remote-image loading with URL policy, byte limits, media checks, and diagnostics
• Font descriptors, deterministic text measurement, TrueType font discovery, and glyph-outline reading
• One shaping-provider contract with a dependency-light managed core-Arabic/TrueType implementation and explicit fallback diagnostics
• Shapes, paths, gradients, shadows, clipping, transforms, vector scenes, and text blocks
• Shared chart kinds, chart snapshots, series data, renderers, and visual-quality reports
• Dependency-free raster buffers, drawing canvases, compositing, patterns, data bars, and sparklines
• First-party PNG/JPEG identification, decoding, encoding, and raster export paths
• Explicit composited-GIF frame selection or animation rejection with typed loss evidence
• Dependency-free TIFF output with uncompressed, PackBits, or Deflate strips and deterministic lossless WebP encoding with common raster export options
• Shared SVG primitive writing and scalable drawing export
• Single and batch image-export builders with dimensions, source metadata, and diagnostics
• Reproducible, output-validated identification, decode, encode, resize, and placement-optimization benchmarks with opt-in library comparisons isolated from runtime packages

Dependency footprint: zero third-party runtime dependencies.

OfficeIMO.Archives

• Bounded ZIP metadata traversal with ingestion warnings and verified member reads in memory
• Managed ZIP extraction with path, declared-size, decoded-length and CRC validation
• Managed single-cabinet extraction with Store, MSZIP, and LZX compression
• Passive MSI extraction maps embedded CAB members through authored Directory, Component, File, and Media tables
• Selective output, path and collision validation, finite resource limits, cancellation, and atomic destination publication

Dependency footprint: only OfficeIMO.Core; extraction does not execute installers or custom actions. Cabinet sets, external MSI payloads, and action-dependent installation layouts are unsupported; see the support matrix.

OfficeIMO.IWork

• Bounded ZIP, directory-bundle, nested Index.zip, Snappy-framed IWA, ArchiveInfo, MessageInfo, and protobuf-envelope reading
• Application detection, producer build history, raster/PDF preview discovery, defensive package-entry access, and preservation of primary and auxiliary IWA payloads
• Extended semantic projections: rich Pages content/layout/images/tables; sparse typed Numbers cells/formulas/merges/table metadata; and positioned Keynote rich text/images/tables/notes
• Opt-in OfficeIMO.Word.IWork, OfficeIMO.Excel.IWork, and OfficeIMO.PowerPoint.IWork adapters with editable reconstruction versus visual fallback reports
• Independently sourced corpus evidence across Pages 14.1/14.5, Numbers 11.1–15.1 histories, and Keynote 8.1/14.5/15.2.1
• Template-free native Keynote slides with rich text, PNG/JPEG images and scalar tables; native Pages body documents with page settings and rich paragraphs; independent decoding and fixture-specific Apple reopen/export/edit evidence
• Template-free native Numbers scalar-table spreadsheets with independent decoding and native open/edit/save/export evidence
• Preservation-aware unchanged native saves, direct Keynote image moves/resizes and existing Numbers numeric/boolean cell edits through separate sessions over immutable source snapshots, with explicit visual-cache acceptance and operation reports

Dependency footprint: only OfficeIMO.Core; the package and IWA readers are first-party implementations. Word, Excel, and PowerPoint remain independent unless an iWork adapter is installed explicitly.

OfficeIMO.Data.Arrow

• Bounded synchronous and asynchronous Apache Arrow record batches from any forward-only DbDataReader
• Native Boolean, numeric, decimal, temporal, GUID, binary, and text arrays with explicit unsupported-type policy
• Shared adapter for OfficeIMO Excel, CSV, and ordinary ADO.NET providers without adding Arrow to their runtime graphs

Dependency footprint: OfficeIMO.Core and Apache.Arrow; Excel and CSV remain independently usable.

OfficeIMO.Data.Generators

• Compile-time RowMapper<T> configuration shared by Excel, CSV, and other DbDataReader sources
• Primary column names, aliases, inherited writable properties, and build diagnostics for unsupported model shapes
• Reflection-free generated mapping validated in a NativeAOT consumer

Dependency footprint: build-time Roslyn analyzer only; no generator assembly is deployed with the application.

OfficeIMO.Drawing.HarfBuzz

• Optional full OpenType GSUB/GPOS shaping through the shared IOfficeTextShapingProvider contract
• Stable logical cluster mappings and positioned glyph advances for TrueType and OpenType/CFF fonts
• Windows, Linux, macOS, and WebAssembly native assets kept out of the dependency-light Drawing and PDF cores

Dependency footprint: OfficeIMO.Core, HarfBuzzSharp, and its platform-native runtime assets.

OfficeIMO.Drawing.CodeGlyphX

• Optional typed bridge from CodeGlyphX QR, matrix, and linear barcode symbols to reusable OfficeDrawing scenes
• Neutral SVG handoff without making either core package depend on the other
• Searchable barcode label text and explicit unsupported-import counts

Dependency footprint: only OfficeIMO.Core and CodeGlyphX; both core packages remain independently usable.

OfficeIMO.ChartForgeX

• Optional bridge from every ChartForgeX VisualArtifact to reusable SVG and OfficeDrawing representations
• Vector-native placement in Word, Excel, and PowerPoint plus first-party PDF composition through OfficeIMO.Drawing
• Point-normalized sizing, accessibility text, region metadata, and explicit preserve-vector, raster-fallback, or require-vector policy
• Native editable Visio projection for CFX topology, flow, and sequence semantics, including containers, Shape Data, hyperlinks, messages, notes, activations, and fragments
• ChartForgeX watermarks and render options flow through the same conversion, while document/page watermarks remain owned by OfficeIMO

Dependency footprint: ChartForgeX and the OfficeIMO Word, Excel, PowerPoint, PDF, Visio, and Core packages. Existing format packages remain independently usable and do not depend on ChartForgeX.

OfficeIMO.ChartForgeX.Markdown

• Optional static Mermaid fence rendering for Markdown-to-Word, PDF, HTML, and PowerPoint pipelines
• Embedded PNG images for documents and isolated SVG images for HTML, with captions, accessibility text, diagnostics, and source fallback

Dependency footprint: ChartForgeX.Markup.Mermaid and OfficeIMO.MarkdownRenderer. The document and Markdown cores remain independently usable.

OfficeIMO.Word

• Create, load, edit, append, inspect, and save .docx documents
• Read, write, and convert the supported first-party Word 97–2003 .doc subset with loss preflight
• Rich runs, fonts, colors, highlights, borders, shading, tabs, spacing, line breaks, and custom paragraph styles
• Bullets, numbering, picture bullets, nested lists, start values, cloning, and list-style detection
• Tables with styles, borders, cell margins, merge/split, nested tables, repeated header rows, widths, heights, and page-break control
• Images from files, streams, bytes, Base64, and URLs with alt text, crop, transparency, wrapping, rotation, flipping, and positioning
• Native charts, shapes, lines, text boxes, equations, embedded documents/objects, and SmartArt inspection/mutation helpers
• Fields, TOCs, bookmarks, hyperlinks, cross-references, document variables, bibliography sources, and field-update reports
• Sections, page sizes, orientation, margins, columns, page/background color, watermarks, and page numbers
• Default, first-page, and even-page headers and footers, including multi-section inheritance and cleanup
• Footnotes, endnotes, comments, revisions, tracked-change helpers, comparison/redline reports, and document merging
• Content controls for text, checkboxes, dates, lists, pictures, rich text, and repeating sections
• Mail merge, formatting-preserving field replacement, conditional template blocks, Custom XML binding, and form-map validation
• Macro add/extract/remove, document protection, encrypted packages, digital-signature inspection, cleanup, repair, and feature preflight
• Managed document export to PNG, JPEG, TIFF, lossless WebP, and SVG; opt-in conversion packages add PDF, HTML, Markdown, RTF, ODT, and Google Docs

Dependency footprint: Open XML SDK plus OfficeIMO.Core; legacy .doc support and image export are OfficeIMO implementations. Apple Pages import is opt-in through OfficeIMO.Word.IWork.

OfficeIMO.Excel

• Create, load, edit, inspect, and save .xlsx workbooks
• Read, write, and convert the supported first-party BIFF8 .xls subset with loss preflight
• Worksheets, cells, range algebra, merges, mutable table schemas, complete filter state, freeze panes, hyperlinks, local/workbook names, and named styles
• Object, dictionary, DataTable, DataSet, row, stream, and typed-model import/export with editable-row workflows
• Streaming reads, direct package writers, parallel compute/apply phases, progress, cancellation, and large-workbook controls
• Fonts, fills, borders, alignment, number formats, rich text, themes, row/column sizing, and reusable report styling
• Transactional row/column/cell shifts plus copy/move/transpose with dry-runs, rollback budgets, reference remapping, and package diagnostics
• Data validation, conditional formatting, icon sets, data bars, color scales, allowed-edit range management, ignored-error metadata preservation, and sparkline lifecycles
• Shared A1/R1C1 formula/reference syntax, formula-aware search and state diagnostics, dependency graphs, and a bounded calculation engine for reporting functions
• Charts across common 2-D/3-D, pie, radar, stock, surface, combo, secondary-axis, trendline, dashboard, and native ChartEx scenarios
• Pivot tables with fields, layouts, styles, filters, grouping, calculated fields, shared caches, and native slicer/timeline views
• Native query-backed tables with caller-hosted execution, explicit security policy, bounded transactional refresh, cancellation, and structural remapping
• Templates with marker binding, repeated rows, repeated sheets, optional regions, formatters, image binding, and preflight diagnostics
• Legacy comments plus threaded-comment/person metadata inspection and preservation
• Worksheet/workbook protection, encrypted OOXML packages, document properties, and compatibility validation
• Print areas, page breaks, page setup, and first/odd/even headers and footers with supported images
• Native in-cell images plus feature inspection and relationship-preserving round trips for macros, external links, custom XML, embedded packages, signatures, controls, and unowned imported parts
• Explicit file-backed editing with size/part budgets and deterministic cancellation, without changing established direct-write, streaming-read, or unchanged-package fast paths
• Workbook, worksheet, and range export to PNG, JPEG, TIFF, lossless WebP, and SVG; adapters add PDF, HTML, ODS, and Google Sheets
• Reproducible read, write, edit, package-size, and feature-rich cross-library benchmark suites with output validation and platform provenance

Dependency footprint: Open XML SDK plus OfficeIMO.Core; legacy .xls support and image export are OfficeIMO implementations. Apple Numbers import is opt-in through OfficeIMO.Excel.IWork.

OfficeIMO.PowerPoint

• Create, load, edit, inspect, and save editable .pptx presentations
• Read, author, edit, preserve, encrypt, and convert .ppt, .pot, and .pps through a versioned capability contract and loss preflight
• Slide creation, duplication, deletion, reordering, sections, presentation sizes, layouts, placeholders, and templates
• Text boxes, rich runs, paragraphs, bullets, alignment, spacing, auto-fit, hyperlinks, and theme-aware typography
• PNG/JPEG/SVG pictures from files and streams with crop, replacement, validation, positioning, and effects
• Tables with merges, cell formatting, borders, fills, pagination helpers, and data-driven creation
• Shared chart authoring, data binding, formatting, markers, axes, combo charts, secondary axes, and chart updates
• Auto-shapes, custom geometry, lines, groups, alignment, distribution, grids, sizing, anchors, stacking, and effects
• Backgrounds, gradients, overlays, themes, color transforms, transitions, speaker notes, notes masters, metadata, and media inspection
• Semantic deck plans and reusable executive-summary, chart-story, comparison, screenshot, appendix, architecture, and closing compositions
• Deck preflight and rhythm analysis for density, repetition, long sections, layout balance, and missing closings
• Feature/package inspection, validation, repair, accessibility metadata, SmartArt inspection, and preservation-aware editing
• Encrypted presentation save/load and read-only, stream-backed, detached-load, and explicit-persistence lifecycles
• Slide export to PNG, JPEG, TIFF, lossless WebP, and SVG plus presentation-wide image export; adapters add PDF, HTML, and ODP

Dependency footprint: Open XML SDK plus OfficeIMO.Core; legacy binary support, composition, editing, charting, and managed image export are OfficeIMO implementations. Apple Keynote import is opt-in through OfficeIMO.PowerPoint.IWork.

OfficeIMO.Visio

• Create, load, edit, inspect, and save drawing, template, stencil, and macro-enabled Open XML Visio packages without Visio automation
• Explicit VDX/VSX/VTX XML import/export through the same model, with bounded loading and fidelity reports
• Multi-page documents, page settings, scale, backgrounds, metadata, document settings, and stream/file lifecycles
• Rectangles, ellipses, diamonds, triangles, callouts, custom/master geometry, groups, and shape duplication
• Connectors, connection points, arrows, routing, line jumps, endpoint queries, and topology inspection
• Shape text, text styles, fills, lines, themes, style sheets, Shape Data, hyperlinks, comments, and protection
• Layers, containers, background pages, page instances, and fluent selection/query helpers
• Built-in and learned masters, stencil profiles, master editing, replacement, and migration plans/artifacts
• Flowchart, block, architecture, network, topology, swimlane, org-chart, sequence, timeline, dependency, and graph builders
• Loaded-diagram editing, layout, selection, validation, package checks, desktop compatibility proof, and visual-quality analysis
• Headless PNG, JPEG, TIFF, lossless WebP, and SVG export for individual pages plus document-wide batch export
• Searchable PDF conversion through OfficeIMO.Visio.Pdf, with explicit semantic-fallback diagnostics

Dependency footprint: System.IO.Packaging plus OfficeIMO.Core; the VSDX model and renderers are first-party, while PDF conversion reuses the shared Reader/PDF projection.

OfficeIMO.Invoicing

Typed invoice and credit-note authoring, parsing, editing and bounded CII/UBL conversion. One decimal calculation owns quantities, adjustments, VAT and payable totals. Unsupported source data is reported before rewriting. The core has no external runtime dependencies.

OfficeIMO.Invoicing.Validation adds hash-pinned XSD and Schematron releases for all five Factur-X 1.09.2 profiles, EN 16931, XRechnung and UBL Peppol BIS, plus pinned independent corpus evidence. Business-rule execution uses an explicitly configured local Java/Saxon runtime. OfficeIMO.Invoicing.Pdf renders localized or multilingual layouts and attaches the exact captured XML through PdfInvoiceDocument. Native Polish FA(3) reading and bounded authoring keep national declarations separate from EN totals, with an offline pinned-schema validator and official fixtures. OfficeIMO.Invoicing.KSeF owns token authentication, encrypted online submission, status and UPO retrieval. Submission references, accepted status, receipt binding and receipt schema results remain distinct.

OfficeIMO.Pdf

• Create PDFs with page setup, rich text, TrueType/OpenType-CFF subsetting, bounded managed Arabic plus shaping-provider positioning, multilingual font fallback, dictionary hyphenation, mixed inline visuals, typed business recipes, page-aware components, styled multipage containers, balanced block-flow columns, tables, and images
• Conditional and replayable flow, position capture, semantic sections, generated TOCs, named destinations, outlines, and generated optional-content layers
• Vector drawings, chart scenes, backgrounds, page decorations, headers, first/even footers, watermarks, metadata, and viewer preferences
• AcroForm creation, field values, choice fields, appearance generation, filling, flattening, and validation
• Annotations, bookmarks/outlines, named destinations, attachments/associated files, optional-content layers, and structured/tagged output
• Exact-artifact validator-backed generation and proof for PDF/A-2b, PDF/A-3b, PDF/UA-1, Factur-X, and ZUGFeRD, plus fail-closed readiness analysis for other formal profiles
• Text extraction by page/range, layout-aware Markdown, logical paragraphs/headings/lists/tables, links, forms, images, and navigation
• Inspect pages, boxes, fonts, images, attachments, outlines, forms, actions, layers, tags, catalog metadata, security, signatures, and revisions
• Extract, split, merge, import, crop, delete, duplicate, reorder, move, rotate, and overlay/underlay complete source pages
• Edit metadata, forms, annotations, bookmarks, attachments, and security; stamp text/images and apply watermarks
• Redaction search/application/verification, sanitization, optimization, OCR hooks, and document-understanding pipelines
• Standard and modern encrypted PDF read/write plus signature mutation and permissions analysis
• Incremental object updates and append-only annotation paths where the source structure allows them
• Managed page rendering to PNG, JPEG, TIFF, lossless WebP, and SVG with page selections, pixel/page limits, capability diagnostics, and continue-on-error batches
• Shared mutation-portfolio and render-compatibility assessments backed by the canonical preflight/planner and generated capability registry
• Bounded stream serialization with per-save peak-retention, spill, buffering, and passthrough evidence
• Exact embedded TrueType outlines plus shared managed CMYK, Lab, XYZ, and calibrated-color conversion where supported
• Logical recovery used by PDF-to-Word, PDF-to-Excel, PDF-to-PowerPoint, and PDF-to-RTF adapters
• Conversion proof, visual comparison, external-validator hooks, and rewrite-preservation reports for warnings, blockers, and structure drift

Dependency footprint: only OfficeIMO.Core; no third-party PDF parser, writer, renderer, OCR, or cryptographic dependency. Install OfficeIMO.Pdf.Ocr only for scanned-page recognition and searchable output, and OfficeIMO.Security only for its built-in CMS/X.509/RFC 3161 adapter.

OfficeIMO.Invoicing.Pdf

• Capture a typed invoice once for visible PDF content and its embedded CII XML
• Render invoices and credit notes with line items, VAT breakdowns, payment details, identifiers, and totals
• Keep later model edits separate from an existing snapshot
• Localize generated labels in English, German, Polish, French, or custom multilingual packs
• Preserve long mixed-script descriptions through embedded font fallbacks and exact XML/PDF agreement checks

Dependency footprint: OfficeIMO.Invoicing and OfficeIMO.Pdf. Install this optional adapter only when combining the XML invoice engine with PDF generation.

OfficeIMO.Pdf.Ocr

• Any IOcrEngine over bounded rendered PDF pages without a Reader or provider dependency
• Standalone raster image recognition at source resolution, with reusable text/table reconstruction and review snapshots for editable format adapters
• Pixel, point, and normalized geometry projection into cropped and rotated visual PDF coordinates
• Confidence and native-text overlap filtering followed by the canonical language-neutral PDF understanding pipeline
• Searchable invisible-text output with per-page provider, model, language, word, rejection, and diagnostic evidence

Dependency footprint: only OfficeIMO.Ocr and OfficeIMO.Pdf; providers and native runtimes remain explicit application choices.

OfficeIMO.Security

• Detached and encapsulated CMS signing and verification with bounded parsing and structured findings
• RSA and ECDSA verification, platform X.509 chain/revocation policy, and RFC 3161 timestamp validation
• CMS EnvelopedData encryption/decryption for S/MIME recipients
• Bounded XML Digital Signature creation, verification, and canonicalization for format-owned signing workflows
• Platform-RSA signing without exporting private keys, including CNG/HSM-compatible key handles
• One strongly typed provider explicitly supplied to thin format-package security adapters

Dependency footprint: OfficeIMO.Core contracts plus BouncyCastle.Cryptography and System.Security.Cryptography.Xml; no dependency on Word, PDF, Email, or another format package. Format packages do not depend on Security.

Format package	Security support without the provider	Optional provider-backed operations
OfficeIMO.Word	OPC/VBA signature discovery, evidence reporting, and fail-safe mutation policy	OPC XML signature creation/validation and VBA CMS/trust/timestamp validation
OfficeIMO.Pdf	Signature dictionaries, byte ranges, external-signer hooks, preservation and mutation policy	Built-in CMS signing, X.509 validation, and RFC 3161 processing
OfficeIMO.Email	MIME parsing, S/MIME carrier discovery, and protected-source retention	CMS signing, encryption, verification, decryption, and sign-then-encrypt for RFC 5322/MIME output
OfficeIMO.Excel	OPC/VBA signature inspection and fail-safe mutation policy	OPC/VBA signature creation, validation, and trust evaluation
OfficeIMO.PowerPoint	OPC, legacy, and VBA signature inspection plus fail-safe mutation policy	OPC/VBA signature creation, validation, and trust evaluation
OfficeIMO.Visio	OPC signature inspection plus fail-safe mutation policy	OPC signature creation and validation
OfficeIMO.OpenDocument	First-party password encryption/decryption, signature discovery, and fail-safe mutation policy	Bounded XML package-manifest signature creation and validation
OfficeIMO.Epub	IDPF/Adobe font deobfuscation, signature discovery, and diagnostics	Bounded XML package-manifest signature creation and validation

Install OfficeIMO.Security only in applications that use a provider-backed operation. Its cryptographic dependencies therefore do not change the restore graph, trimming roots, or NativeAOT surface of ordinary format consumers.

OfficeIMO.OpenDocument

• Native ODT, ODS, ODP, and ODG package and flat-XML loading, editing, inspection, and deterministic saving
• ODT paragraphs, headings, runs, styles, lists, tables, links, bookmarks, sections, page layout, headers/footers, images, and tracked changes
• ODS sparse/repeated cells, typed values, formulas, styles, merges, sizing, visibility, names, validation, and print ranges
• ODP slides, masters/layouts, text, shapes, groups, images, crop, tables, notes, backgrounds, transitions, and basic animation metadata
• ODG pages, groups, basic shapes, text and image editing, FODG conversion, and a reported drawing projection for shared exporters
• Unknown XML and package-part preservation with explicit loss and capability reports
• Dependency-free AES-256-CBC password encryption/decryption with bounded aggregate KDF work and hash-pinned LibreOffice interoperability evidence

Dependency footprint: only OfficeIMO.Core; no OpenDocument SDK and no LibreOffice runtime.

OfficeIMO.Publisher

Native Publisher publication recovery into positioned pages, styled text stories, table cells and embedded image assets, with SVG export and explicit fidelity reports. OfficeIMO.Publisher.Pdf adds multi-page PDF conversion through the shared drawing and PDF engines. The support contract identifies profile and rendering limits. The reader references only OfficeIMO.Core and does not require Publisher. The Reader adapter retains complete stories, physical page inventory and image descriptors for semantic ingestion.

OfficeIMO.Chm

Compiled HTML Help reading with exact entry extraction, topics, embedded resources, hierarchical contents, and the keyword index. Optional Markdown, EPUB, and PDF adapters reuse the existing conversion engines and expose fidelity reports. Reader ingestion retains archive hashes and topic citations. See supported profiles for ITSF/LZX coverage and export limits. Loading is inert and needs no Windows help viewer.

OfficeIMO.DjVu

Managed DjVu scanned-book reading, stored UTF-8 text and word geometry, bounded page rendering, and source navigation. OfficeIMO.DjVu.Pdf adds searchable scanned-page PDFs and explicit OCR for missing text; OfficeIMO.Reader.DjVu projects the native document into Reader. See the codec and independent qualification contract.

OfficeIMO.Xps

Portable XPS/OpenXPS package reading, native page editing, and writing, with embedded fonts, paths, images, and explicit conversion diagnostics. SVG and raster export reuse the shared drawing engine; OfficeIMO.Xps.Pdf adds vector PDF export. Rendering and preservation boundaries are documented separately. OfficeIMO.Reader.Xps adds bounded Unicode, logical-order and page ingestion. The core package references only OfficeIMO.Core.

OfficeIMO.Rtf

• Bounded RTF lexer/parser with a lossless syntax tree and exact unchanged-source round trips
• Editable semantic model for paragraphs, runs, styles, lists, tables, sections, headers/footers, notes, fields, images, shapes, objects, comments, and revisions
• Canonical and preserve-mode writing with structured parser, binding, and conversion diagnostics
• HTML bridge and dedicated Markdown, PDF, and Word workflow adapters

Dependency footprint: System.Text.Encoding.CodePages plus OfficeIMO.Core; no third-party RTF parser.

OfficeIMO.Markdown

• Typed Markdown AST and fluent builder for headings, paragraphs, lists, tasks, tables, code, callouts, details, definitions, front matter, footnotes, TOCs, and semantic fenced blocks
• Native parsing with source spans, anchors, stable block identities, transforms, and diagnostics
• HTML fragment/document rendering with CSS profiles and optional Prism, Mermaid, chart, and math shell assets
• AOT-friendly typed selectors and DTO-style AST projection for editor, chat, transcript, and document hosts

Dependency footprint: only OfficeIMO.Core; Markdown parsing and writing are first-party.

OfficeIMO.Adf

• Lossless Atlas Document Format JSON model with unknown nodes, marks, attributes, and extension properties preserved
• Structural validation plus Markdown and HTML projections with explicit fidelity diagnostics
• Markdown and HTML import through OfficeIMO's existing document engines

Dependency footprint: OfficeIMO Markdown, Markdown.Html, and HTML plus System.Text.Json on compatibility targets; no Atlassian SDK.

OfficeIMO.Html.Core

• Owned HTML nodes, attributes, document modes, source offsets and separate template contents
• Immutable snapshots, independent edits, stable node IDs and detached-node ownership
• Parser, selector/serialization and charset contracts without graphics or browser dependencies

OfficeIMO.Html.AngleSharp

• Inert full-document parsing with source limits and cooperative cancellation
• Selectors and serialization over owned nodes, preserving parser recovery during edits
• Existing web encoding aliases and code-page support behind a replaceable provider

OfficeIMO.Html

• Canonical HtmlConversionDocument with DOM, base-URI, media, resource, and URL-policy ownership
• CSS-aware layout scene shared by PNG, JPEG, TIFF, SVG, WebP, PDF, and Office adapters
• Direct PNG, JPEG, TIFF, SVG, and lossless WebP output with structured diagnostics and bounded local/remote resource loading
• Bounded CSS math, deterministic media preferences, caller stylesheets, paged running strings, shared CSS/SVG color parsing, WOFF 1, and Unicode-range-aware font selection
• Stable HTML contracts reused by Office, Markdown, Reader, PDF, and optional cross-format bridges

Dependency footprint: OfficeIMO.Core, AngleSharp, and AngleSharp.Css. Email, RTF, MHTML, and PDF are not part of the base HTML restore graph.

OfficeIMO.Html.Rtf

• Semantic HTML-to-RTF and RTF-to-HTML conversion with structured fidelity diagnostics
• Existing OfficeIMO.Html conversion namespaces retained while package ownership becomes explicit

Dependency footprint: OfficeIMO.Core, OfficeIMO.Html, and OfficeIMO.Rtf.

OfficeIMO.Mhtml

• MHTML/MHT loading and deterministic saving with HTML root selection
• CID and Content-Location resource resolution through the shared HTML resource policy

Dependency footprint: OfficeIMO.Core, OfficeIMO.Html, and OfficeIMO.Email for MIME parsing.

OfficeIMO.Email.Html

• One safe HTML/RTF/text body-selection contract with untrusted sanitization and remote resources blocked by default
• CID, Content-Location, absolute-location, and filename resource resolution with bounded operation-scoped reads
• Shared prepared projection consumed by OfficeIMO.Email.Image and OfficeIMO.Reader.Email

Dependency footprint: OfficeIMO.Email, OfficeIMO.Html, and OfficeIMO.Html.Rtf; the base Email package remains HTML-free.

OfficeIMO.Email.Image

• Email body export through the HTML image pipeline with plain-text and RTF fallbacks
• Inline MIME/CID resources, page selection, diagnostics, and bounded output

Dependency footprint: OfficeIMO.Core, OfficeIMO.Email, OfficeIMO.Email.Html, and OfficeIMO.Html.

OfficeIMO.Mhtml.Pdf

• Bounded MHTML-to-PDF conversion with embedded MIME resources and combined diagnostics
• First-party PDF document, bytes, stream, path, sync, and async result paths

Dependency footprint: OfficeIMO.Core, OfficeIMO.Mhtml, OfficeIMO.Html.Pdf, and OfficeIMO.Pdf.

OfficeIMO.AsciiDoc

• Dependency-free, source-preserving AsciiDoc parser, typed tree, semantic model, and writer
• Headings, paragraphs, lists, definitions, admonitions, delimited blocks, tables, images, anchors, attributes, and STEM content
• Preserve and canonical output modes with source-located diagnostics
• Explicit bounded processing with root-confined include policy; parsing never executes directives

Dependency footprint: only OfficeIMO.Core; no Asciidoctor process or parser package.

OfficeIMO.Latex

• Source-preserving parser for a bounded LaTeX2e interoperability profile
• Tokens, groups, commands, environments, comments, headings, lists, figures, tables, labels, references, citations, theorems, and math
• Exact unchanged-source writing and visible preservation of unknown commands/environments
• Opt-in bounded expansion for safe document-local simple macros

Dependency footprint: only OfficeIMO.Core; no TeX runtime, compiler, or parser dependency.

OfficeIMO.Bibliography

• Format-neutral citation keys, item kinds, contributors, dates, identifiers, publication fields, keywords, notes, and ordered native extensions
• Source-preserving read, edit, deterministic write, and reopen workflows for BibTeX/BibLaTeX, CSL JSON, RIS, NBIB/MEDLINE, and EndNote XML
• Exact unchanged text and loaded-byte output, plus safe same-format unknown-field preservation after edits
• Cross-format conversion reports with strict rejection of approximated or omitted data
• Bounded file, stream, text, sync, and async APIs with cancellation and inert XML/TeX handling

Dependency footprint: System.Text.Json and System.Text.Encoding.CodePages on compatibility targets; no dependency on another OfficeIMO package, Word, Open XML, TeX, EndNote, rendering, or network clients.

OfficeIMO.Opml

• OPML 1.0/2.0 create, read, edit, validate, convert, and deterministic write
• Nested outlines, subscription attributes, qualified extension attributes, and exact unchanged-source output
• Explicit size, character, depth, outline, and attribute limits with DTD processing disabled

Dependency footprint: only OfficeIMO.Core; no external parser or runtime dependency.

OfficeIMO.Project

• Microsoft Project XML create, read, edit, validate, and source-preserving save
• Qualified MPP/MPT 8, 9, 12, and 14 authoring and editing, document templates, MPX 4 exchange, and assessed format conversion
• Typed tasks, calendars, resources, assignments, baselines, custom fields, and thin fluent builders
• Explicit assignment/progress/cost calculation, resource leveling, earned value, bounded formulas/lookups, and caller-resolved external schedules
• Portable project views and mapped tables; optional PDF/HTML/image and editable Office reports through OfficeIMO.Workflows
• Explicit input limits, cancellation, stable identities, and loss assessment

Native producer profiles, opaque-record preservation, scheduling, legacy formats, and rendering have separate operation boundaries.

Dependency footprint: only the matching coordinated OfficeIMO.Core release; no COM, external parser, or network dependency in file operations.

OfficeIMO.DocBook

• DocBook 4.5 and 5.2 article/book creation, bounded reading, editing, validation, conversion, and writing
• Typed common structures for metadata, sections, lists, tables, code, links, notes, media, and indexes
• Lossless extension preservation and exact schema identifiers without claiming exhaustive vocabulary validation

Dependency footprint: only OfficeIMO.Core; external schemas are identified but not downloaded at runtime.

OfficeIMO.CSV

• First-class headers/rows document model with file, stream, text, in-memory, and forward-only streaming lifecycles
• Single- and multi-character delimiters, culture, encoding, newline, quote, escape, whitespace, comment, and null-token controls
• Duplicate/blank header policy, W3C #Fields: support, static metadata columns, row-length policy, and custom date formats
• Gzip, deflate, Brotli, and zlib CSV read/write with extension-based detection
• Add/remove/transform columns and rows, filter, sort, materialize, and culture-aware save workflows
• Schema inference and validation with required/optional typed columns, defaults, conversion delegates, and custom rules
• Reflection-free typed object mapping suitable for trimming and NativeAOT-sensitive consumers
• DataTable, IDataReader/DbDataReader, typed-reader, SQL/bulk-copy-shaped, reusable-row, field-span, and trusted-text paths
• Ordered bounded parallel row projection plus a .N