dsh-attachment-formats
DeepSeek Harness web plugin that expands composer attachments with PDF, Office, OCR, text, and image conversion workflows.
Install
$ dsh plugin --profile web add github:linkingoscar/dsh-attachment-formatsPlugin Overview & Capabilities
AI-assisted organization based on the public repository snapshot. The content must be grounded in source evidence and does not replace compatibility or security verification.
Key Capabilities
- Adds file-picker, drag-and-drop, and paste workflows for supported attachment formats.
- Extracts text from PDFs, DOCX, XLSX, PPTX, plain text, Markdown, JSON, and code files.
- Uses OCR for scanned PDFs, with optional Baidu OCR, remote vision-language OCR, and local tesseract.js paths.
- Converts TIFF and browser-decodable image formats to PNG for the native image draft rail.
- Shows text attachments as removable document cards and merges their content when sending.
- Spills long documents to `.dsh-attachments` workspace storage and supplies index cards for model-assisted reading.
- Provides `/attach list` and `/attach full <id|name>` commands for spilled documents.
- Offers an attachment-cache settings page and optional workspace zero-copy references for eligible text files.
Useful For
- Attach and discuss PDF, Office, and text documents in the DeepSeek Harness web composer.
- Extract text from scanned PDFs or fall back to page images for vision-capable models.
- Keep oversized documents out of the prompt while allowing the model to inspect workspace content with reading tools.
- Convert legacy Office files when LibreOffice is available.
- Handle drag-and-drop or pasted mixed attachment batches in the web UI.
Who It Fits
- DeepSeek Harness web users who work with document attachments.
- Users analyzing PDFs, Office documents, and scanned files in chat workflows.
- Developers who want attachment conversion without modifying DeepSeek Harness core packages.
Documented Limitations
- OCR quality is limited for low-resolution scans and complex tables; low-confidence OCR falls back to page images.
- The high-fidelity pymupdf4llm PDF path handles PDFs of up to 40 pages; larger files use pdfjs.
- Legacy `.doc`, `.xls`, and `.ppt` require LibreOffice; RTF requires pandoc.
- DOCX formulas and embedded images are not extracted; XLSX charts and comments are not extracted.
- iWork files and archives are not supported.
- Image synthetic drops may also be accepted by other idle conversations when several conversations are open, due to a harness-level behavior.
- Document-card merge relies on a DOM-event bridge to an unpublished harness API and may break after core upgrades.
DSH Compatibility
Version-specific runtime evidence collected by DSH Plugin. A missing result means we have not tested that combination yet.
Security Signals
Objective signals discovered from package metadata and source inspection. These are not a guarantee that a plugin is safe.
package.json declares 9 runtime, 5 development, 0 peer, and 0 optional dependencies.
package.json declares DSH bundle metadata.
GitHub reports the repository license as Apache-2.0.
A root package.json was captured and can be inspected by the registry.
Public GitHub source metadata is available for this registry snapshot.
Source & Registry Notes
Public provenance, Registry classification, and the latest source check for this entry, kept separate from runtime verification.
- Source repository
- genusamblyrhynchusbrunooftoul602/dsh-attachment-formats
- Registry source
- GitHub · dsh-plugin topic
- Registry classification
- Plugin
- Source checked
- 345b112 · 2026-09-20
This project is independently indexed from public source information. DSH Plugin is not affiliated with DeepSeek or the plugin author. Always check the author repository before installation.
Repository Activity
- GitHub stars
- 0GitHub stars
- Forks
- 0
- Open issues
- 0
- Last commit
- 2026-09-19
- Last release
- No release detected
Related DSH Plugins
Ranked by overlapping capabilities, use cases, plugin type, categories, and DSH profile.
Compatibility bridge and host ABI enabling unmodified Pi plugins to run natively on DeepSeek Harness.
View plugin →Persistent multi-model Agent teams and bounded DAG workflows with GUI controls for DeepSeek Harness.
View plugin →Preview, create, and edit spreadsheets, docs, slides, and canvases in DeepSeek Harness powered by Univer.
View plugin →The most comprehensive DeepSeek Harness plugin market — refreshed daily, sourced across the Internet, reviewed before publishing.
View plugin →