Models & MCPPlugin

dsh-memory

FuRongJun-1999/dsh-memory

dsh-memory is a tool for DeepSeek Harness. White-box AGI architecture exploration: metacognition (self-awareness loop), continual learning (knowledge flywheel), world model (condition space + semantic spatiotemporal graph), self-improvement (bootstrapping discipline), zero-LLM white-box pipeline and auditable trust guardrails.

View repository

FuRongJun-1999/dsh-memory Screenshot preview

Stars: 431
Forks: 51
Issues: NaNm
Update: 1 day

REVIEW SCORECARD

Review Scorecard

8.1/10Good

License compliance9.0

Engineering quality8.0

Content completeness6.0

Repository scale7.0

Maintenance activity9.0

✓ No risk detected · Automated scan results are for reference only, not an official endorsement

AI DEEP REVIEW

Worth AI Review

8.0/10Recommended

A well-documented, actively maintained persistent-memory MCP plugin with reproducible benchmarks and honest limitations;

Code quality8

README documents a Rust retrieval kernel with zero third-party deps, a Python MCP server, TypeScript build via tsc, and public reproducible eval scripts (md_cg/eval_common.py, bench_locomo_zh_public), though no CI config or test suite is visible in the excerpt.

Security7

Write access is disabled by default (read-only 'guest' until credentials are configured) and the README emphasizes deterministic rule-based gating plus audit trails, but the credential mechanism itself is not described in the excerpt, so risk cannot be fully assessed.

Utility9

Solves the common agent pain point of cross-session memory loss with a plain-markdown cognitive graph, MCP compatibility across DSH/CodeBuddy/ZCode/Codex/Claude Code, and a three-step install.

Maintenance8

Last updated 2026-09-10, not archived, versioned at v0.4.5 with an active benchmark/report cadence and 159 stars indicating ongoing development.

Documentation8

README provides install commands, cordis.yml config example, tool-chain behavior table, prerequisites, and links to detailed docs, though the full 30+ config options live in a separate file not shown here.

Generated by AI after reading the project README — a decision aid, not an endorsement. Neutral scores are given when information is limited.

PROJECT READMEREADMEExpandCollapse

Loading GitHub README...

INSTALL REFERENCE

Install reference

Structure check pending

dsh plugin --profile web add .

This command was extracted from an explicit README install reference and is display-only; the DSH host will not execute it automatically.

PROJECT TOPICS

Project topics

agent-safety agentic-ai agi ai-agent-framework deepseek dsh-plugin explainable-ai guardrails knowledge-graph mcp multi-agent persistent-memory rag spatiotemporal-graph-neural-network

VALIDATION LADDER

Validation progress

Structure check pending

01Store discoveryPassed

02ClassificationPassed

03Structure checkPending

04Sandbox validationPending

FAQ

Frequently Asked Questions

What is dsh-memory used for?

White-box AGI architecture exploration: metacognition (self-awareness loop), continual learning (knowledge flywheel), world model (condition space + semantic spatiotemporal graph), self-improvement (bootstrapping discipl

How popular is dsh-memory?

dsh-memory has 431 stars and 51 forks on GitHub, last updated 2026-10-21.

What license does dsh-memory use?

dsh-memory is distributed under the MIT license.

Is dsh-memory compatible with DeepSeek Harness?

dsh-memory is listed in the DSH Universe directory as a tool for DeepSeek Harness. This plugin is listed in the DSH Universe directory and covered by its validation pipeline.

CLASSIFICATION EVIDENCE

Classification evidence

Project typePlugin

Feature categoryModels & MCP

Rule confidenceMedium

The system checks GitHub Topics against the site taxonomy and keyword rules. Current matches:No valid category tags。

dsh-memory

Review Scorecard

Worth AI Review

Install reference

Project topics

Validation progress

Frequently Asked Questions

Classification evidence

Related projects

dsh-purge

AI-Novel-Writer

agentrq

awesome-ai-plugins

sandbase-harness

dsh-desktop