DeepSeek Harness Plugins
DeepSeek Harness PluginManifest Valid

dsh-image-unlock

DeepSeek Harness plugin that admits image attachments in the Web GUI and converts them to text markers for a vision-skill workflow on text-only opencode-go upstreams.

VisionUI & ProductivityDeveloper ToolsSkills & Workflows
1GitHub Stars0ForksUpdated2026-08-15

Install

$ dsh plugin --profile web add <本插件目录>

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-image-unlock is an installable DeepSeek Harness bundle that patches opencode-go default input modalities and manages image textification at runtime. It is intended to let attachments pass the GUI/API gate while routing actual image recognition through a compatible vision skill.

Key Capabilities

  • Adds a Cordis bundle patch that sets opencode-go default input to text and image for models without an explicit input declaration.
  • Patches dsh-llm-pi-ai serialization to convert durable image blocks, including nested tool-result blocks, into attachment-path text markers on the opencode-go route.
  • Controls textification through the presence of a claude-vision-skill directory, with configuration overrides for textification and skill location.
  • Watches the skills directory so skill-driven mode changes take effect without restarting.

Useful For

  • Use image attachments with text-only opencode-go models without the Web GUI rejecting them.
  • Pass attachment paths to a vision-skill workflow for image recognition.
  • Switch between textified vision-skill handling and native image parts when using an appropriate visual model.

Who It Fits

  • DeepSeek Harness Web GUI users
  • Users of opencode-go-backed models
  • Developers maintaining a claude-vision-skill workflow

Documented Limitations

  • The plugin does not make a text-only model capable of understanding images; actual recognition requires a vision skill or a native vision model.
  • Image textification affects only the opencode-go route; other pi-ai routes are unchanged.
  • Its dsh-llm-pi-ai node_modules patch can be overwritten by pnpm install or DSH upgrades and must then be rerun.
  • Installation requires an additional profile cordis.patch.yml entry and a one-time code patch according to the README.

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, 0 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
FrostLeafKEE/dsh-image-unlock
Registry source
GitHub · dsh-plugin topic
Registry classification
Plugin
Source checked
f4e51d8 · 2026-08-22

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
1GitHub stars
Forks
0
Open issues
0
Last commit
2026-08-15
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