DeepSeek Harness PluginManifest Valid

dsh-file-attachment

Drag, paste, or upload files and images into the DSH Web GUI with inline previews and VLM descriptions.

Terminal & TUIDeveloper ToolsSecurity & PolicyRemote Execution

3GitHub Stars0ForksUpdated2026-10-26

Source

Install

$ dsh plugin --profile web add @wszhoho/dsh-file-attachment

Plugin 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.

source-grounded

dsh-file-attachment is an installable DeepSeek Harness (DSH) Web GUI plugin that enhances chat attachments. It allows users to drag and drop, paste from clipboard, or use the paperclip button to upload multiple documents, code files, and images. Attachments are saved to the local workspace and injected as absolute path references for models, with clickable image lightboxes and optional OpenAI-compatible VLM descriptions for text-only models.

Key Capabilities

Support for drag-and-drop, clipboard paste, and native paperclip button multi-file uploads
Unified workspace persistence pipeline under .dsh-file-attachment with automatic deduplication and conflict resolution
Automatic insertion of @absolute-path references directly into the chat composer
Inline attachment previews in composer and clickable thumbnail lightboxes in chat history
Optional VLM integration to automatically generate Chinese image descriptions for text-only models and describe_image tool registration

Useful For

Uploading local documents, configurations, and source code into conversation context via path chips
Pasting and previewing screenshots or diagrams directly in the DSH Web UI
Enabling text-only DeepSeek models to understand uploaded images via automated VLM description backfilling

Who It Fits

DeepSeek Harness Web GUI users working with multi-file workflows and screenshots
Developers needing seamless local file path insertion in conversations
Users running text-only LLMs who want automated multimodal vision descriptions

Documented Limitations

Enforces a strict 50MB single-file size limit
Directory drag-and-drop is not supported
VLM image recognition requires configuring an external OpenAI-compatible multimodal endpoint

DSH Compatibility

Version-specific runtime evidence collected by DSH Plugin. A missing result means we have not tested that combination yet.

Not tested yetNo runtime compatibility tests have been published yet.

Security Signals

Objective signals discovered from package metadata and source inspection. These are not a guarantee that a plugin is safe.

Dependency Counts

package.json declares 0 runtime, 0 development, 2 peer, and 0 optional dependencies.

info

Dsh Bundle Declared

package.json declares DSH bundle metadata.

info

License Declared

GitHub reports the repository license as MIT.

info

Package Manifest Available

A root package.json was captured and can be inspected by the registry.

info

Source Available

Public GitHub source metadata is available for this registry snapshot.

info

Source & Registry Notes

Public provenance, Registry classification, and the latest source check for this entry, kept separate from runtime verification.

Source repository: wszhoho/dsh-file-attachment
Registry source: GitHub · dsh-plugin topic
Registry classification: Plugin
Source checked: 32f64f5 · 2026-10-27

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.