DeepSeek Harness Plugins
DeepSeek Harness PluginManifest Valid

dsh-image-subagent

Enables image attachments in text-only DSH model routes by projecting attachments into rich text placeholders for a vision subagent.

VisionDeveloper Tools
4GitHub Stars0ForksUpdated2026-08-21

Install

$ dsh plugin --profile web add ./dsh-image-subagent

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
A DSH plugin for workflows where a text-only primary model delegates image reading to a configured vision subagent using the read_image tool. It patches route modality declarations so image attachment admission can pass without sending image data directly to text-only endpoints.

Key Capabilities

  • Bundles a Cordis patch for DeepSeek Harness.
  • Allows image attachments to pass the DSH 0.1.1+ paste/admission gate for text-only routes by omitting their inputModalities declaration.
  • Projects attachments into rich text placeholders containing the full attachment ID and object path.
  • Supports delegation from the primary text model to a vision-capable subagent using read_image.
  • Leaves routes that explicitly declare image input unaffected.

Useful For

  • Using a text-only primary model while analyzing pasted image attachments through a vision subagent.
  • Providing a primary model with an attachment reference it can pass to a configured image-reading subagent.
  • Maintaining native direct image handling for vision routes while adding an indirect image workflow for text-only routes.

Who It Fits

  • DeepSeek Harness users running the web profile.
  • Users with a configured vision subagent and read_image access.
  • Teams using text-only primary model routes alongside vision-capable subagents.

Documented Limitations

  • Version 0.2.0 requires @deepseek-ai/dsh >= 0.1.1-rc.2 and Node.js >= 20.
  • A vision subagent must declare image input and include the read_image tool.
  • The primary text model does not read images directly; it must delegate using the placeholder-provided attachment path.
  • Installation instructions state that pnpm is required and that dsh web must be restarted after installation.

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, 1 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
yuqingsh/dsh-image-subagent
Registry source
GitHub · dsh-plugin topic
Registry classification
Plugin
Source checked
289fdd8 · 2026-09-13

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
4GitHub stars
Forks
0
Open issues
0
Last commit
2026-08-21
Last release
No release detected
For maintainers

Maintaining this plugin?

This listing is generated from public repository data. If you maintain this project, you can review the information and share this listing with your users if you find it useful.

Add to README
Listed on DSHPlugin.app