DeepSeek Harness Plugins
DeepSeek Harness PluginManifest Valid

dsh-ocr1-memory

DeepSeek-OCR optical compression memory plugin that renders text memories into images with Set-of-Mark segmentation and age-based resolution decay.

VisionTerminal & TUIDeveloper Tools
2GitHub Stars0ForksUpdated2026-09-16

Install

$ dsh plugin --profile web add github:DDDFXYqiming/dsh-ocr1-memory

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-ocr1-memory is a DeepSeek Harness plugin inspired by DeepSeek-OCR (OCR1). It segments and renders text memories into Set-of-Mark (SoM) images, applying multi-tier resolution decay (vivid, normal, fuzzy) based on memory age. The plugin supports active recall to restore resolution upon retrieval, optional optical locator segment selection, deterministic text retrieval, and multi-tier memory governance (L1/L2/L3) integrated into DSH context flows.

Key Capabilities

  • Render text memory paragraphs into Set-of-Mark (SoM) visual images
  • Three-tier resolution aging decay (vivid, normal, fuzzy) with active recall restoration
  • Optical locator support via OpenAI-compatible endpoints for top-K segment selection
  • Full governance tool suite for L1/L2/L3 memory read, write, rollback, archive, and indexing
  • Automatic pending candidate generation on turn end and scheduled maintenance tasks
  • Optional visual embedding evaluation and llama-server lifecycle management

Useful For

  • Compressing large context histories into optical representations to reduce token overhead
  • Long-term agent memory retention with age-based detail degradation and targeted recall
  • Structured memory governance with automatic pending distillation and namespace management

Who It Fits

  • DeepSeek Harness agent developers building long-running memory architectures
  • Researchers exploring multimodal optical context compression techniques
  • Developers seeking hierarchical L1/L2/L3 memory persistence for DSH profiles

Documented Limitations

  • Full optical retrieval and embedding features require an OpenAI-compatible multimodal endpoint or local llama-server backend
  • Optical locator accuracy depends on backend model compliance with binary tag extraction rules
  • Automatic server management requires configured paths to local binaries and model directories

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-ocr1-memory
Registry source
GitHub · dsh-plugin topic
Registry classification
Plugin
Source checked
1e01c36 · 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
2026-09-16
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