dsh-memory
Run the following command in DeepSeek Harness:
dsh plugin install boomzikazita/dsh-memory
Paste the following prompt into your AI chat to install this plugin:
Install this plugin in DeepSeek Harness by running dsh plugin install boomzikazita/dsh-memory; the full source is available at https://github.com/boomzikazita/dsh-memory .
About this plugin
Every DeepSeek Harness session is stateless; close the terminal and the context is gone. dsh-memory adds a persistent memory layer so the model still remembers confirmed rules, corrected preferences, and distilled skills across days and projects.
The write path is proposal-based: memory_suggest enqueues a candidate, budget checks and sensitive-pattern screening run, and only memory_confirm promotes it into L1. At recall time, five RRF channels - rule files, session full-text index, semantic search, wiki, and time-window - are fused into a single ranked evidence ledger with per-item trust labels. Secrets are isolated in a Fernet-encrypted vault; plaintext never appears, only vault references. A session-transcription module incrementally indexes logs and extracts correction signals for skill distillation.
Built for developers who collaborate with dsh long-term, manage multiple rule sets and credentials, and want to turn ad-hoc experience into reusable, auditable skills.
Use Cases
- Retain confirmed user preferences and project rules across sessions
- Securely store and retrieve API keys and sensitive credentials in an encrypted vault
- Extract correction signals and reusable skills from historical session logs
Best For
- Developers collaborating with dsh on long-running multi-round projects
- Tech leads needing centralized management of multiple rule sets and credentials
- Engineers who want to turn ad-hoc experience into auditable, reusable assets
Related Plugins
Traceable, searchable cross-session memory for AI agents that turns conversation knowledge into a typed knowledge graph and recalls relevant subgraphs instead of replaying full history, natively integrated with DeepSeek Harness.
Gives DSH AI cross-session long-term memory, to-do and skill management, plus multi-session orchestration, external AI delegation, and an infinite canvas that grows with you.
dsh-mnemon is a three-tier, pluggable, Agent-driven memory system for DeepSeek Harness, combining Runtime memory, Project Documents, and replaceable Memory Spaces with nine long-term providers.
An opinionated, zero-infrastructure file-based memory protocol: human-readable markdown files plus a bounded always-loaded index, curation discipline, and no database or embeddings.