← DeepSeek Harness Plugins
DeepSeek Harness PluginManifest Valid

dsh-ocr1-memory

DeepSeek-OCR based optical context compression memory plugin for DeepSeek Harness.

VisionTerminal & TUIDeveloper Tools
2GitHub Stars0ForksUpdated2026-10-15

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 that implements an optical context compression memory architecture based on DeepSeek-OCR (OCR1). It renders segmented long-context text into Set-of-Mark (SoM) labeled images, utilizing high-density visual tokens instead of raw text tokens to compress agent memory. It features dynamic temporal resolution decay (vivid, normal, fuzzy), active recall restoration, visual locator filtering, and compression metrics tracking.

Key Capabilities

  • Renders text into Set-of-Mark (SoM) annotated images for optical context compression
  • Manages multi-tier temporal decay across vivid, normal, and fuzzy resolutions
  • Triggers active recall to restore compressed memory back to high fidelity upon query
  • Integrates an optional local GGUF/llama-server optical locator for fast candidate filtering
  • Exposes agent tools: ocr1_mem_store, ocr1_mem_retrieve, ocr1_mem_metrics, ocr1_mem_update, ocr1_mem_calibrate, and ocr1_mem_forget
  • Tracks token savings and optical compression ratios with ocr1_mem_metrics

Useful For

  • Compressing large agent interaction logs and long-term memory into visual token representations
  • Cost-effective archival and retrieval of long contextual history in DeepSeek Harness
  • Managing aging memory lifecycle with automatic resolution degradation and retrieval-based revival

Who It Fits

  • DeepSeek Harness developers building long-running autonomous agents
  • Engineers optimizing context window token budgets and memory storage costs
  • Researchers exploring multimodal optical context compression

Documented Limitations

  • Local optical locator and automatic server startup require external binary dependencies such as llama-server and compatible GGUF models
  • Requires DeepSeek-OCR compatible backend services for full OCR reconstruction workflows
  • Requires Node.js >= 22.19 runtime environment

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, 4 peer, and 0 optional dependencies.

info
Dsh Bundle Declared

package.json declares DSH bundle metadata.

info
License Declared

GitHub reports the repository license as BSD-3-Clause.

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
DDDFXYqiming/dsh-ocr1-memory
Registry source
GitHub · dsh-plugin topic
Registry classification
Plugin
Source checked
c0ef660 · 2026-10-22

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
2GitHub stars
Forks
0
Open issues
0
Last commit
2026-10-15
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