DeepSeek Harness Plugins
DeepSeek Harness PluginManifest Valid

dsh-vision-skill

Standard vision skill and toolset for DeepSeek Harness providing image analysis, OCR, visual grounding, and paste-to-path handling.

VisionUI & ProductivityTerminal & TUIDeveloper ToolsSkills & Workflows
2GitHub Stars0ForksUpdated2026-09-18

Install

$ dsh plugin --profile web add github:DDDFXYqiming/dsh-vision-skill

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-vision-skill is a native DeepSeek Harness plugin providing a multi-modal vision runtime skill and eight specialized tools. It integrates Qwen dynamic-resolution preprocessing, multi-provider VLM failover, progressive tool exposure, local Tesseract-first long-screenshot OCR, and zero-patch paste-to-path image delivery to enable text-only models to process visual inputs securely and efficiently.

Key Capabilities

  • Comprehensive image analysis across multiple modes (general, OCR, table, code, error, structured evidence)
  • Visual grounding and object detection returning normalized and pixel bounding boxes
  • Local dominant color extraction without external API calls
  • Chunked long-screenshot OCR using local Tesseract with VLM fallback
  • Paste-to-path web client injection saving pasted images directly to the workspace
  • Clipboard screenshot capture and local caching via SHA-256 hashes
  • Progressive tool exposure to minimize agent context overhead
  • Multi-provider failover with automatic 429 rate-limit backoff and path fencing security

Useful For

  • Analyzing diagrams, error screenshots, code snippets, and structured tables inside DSH workflows
  • Extracting text from long vertical screenshots or chat logs using hybrid OCR
  • Locating specific UI elements or coordinates for UI automation and visual inspection
  • Pasting images directly into the DSH web chat interface without image payload errors

Who It Fits

  • DeepSeek Harness developers working with text-only models needing vision capabilities
  • DSH agent builders requiring visual grounding and UI element detection
  • Users needing automated OCR and document analysis within DSH profiles

Documented Limitations

  • Requires Python runtime with supporting libraries (Pillow, optional Tesseract) for local execution scripts
  • Requires configured OpenAI-compatible multi-modal API endpoints and valid credentials for VLM operations
  • May conflict with user-level or project-level skills if identically named 'vision'

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.

No automated security signals have been published yet.

Source & Registry Notes

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

Source repository
DDDFXYqiming/dsh-vision-skill
Registry source
GitHub · dsh-plugin topic
Registry classification
Plugin
Source checked
6fe7fc1 · 2026-09-19

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
2GitHub stars
Forks
0
Open issues
0
Last commit
2026-09-18
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