Memory

What it is — Hindsight provides long-term project memory management for coding agents.

Who it is for — Developers using AI coding assistants such as Claude Code for code development should install it. If you use a non-Python language for the agent, you can choose other implementations.

Watch out — The plugin requires an API key or token and will execute shell commands. Sandbox testing has not been performed. No obvious pitfalls found.

The verdict — Without sandbox testing results, I would not install it.

What it is — dsh-memory-plugin is an OpenViking memory and context bundle plugin.

Who it is for — When you use the DSH plugin system, people who need OpenViking memory and context bundle are suitable to install dsh-memory-plugin. For people who do not need OpenViking as context database, they can use other plugins.

Watch out — Sandbox test passed, installed in a fresh profile and registered into profile by harness. It will execute shell commands. No obvious pitfalls found.

The verdict — I would install it because it can be registered by harness after installation in a fresh profile.

What it is — The plugin integrates DeepSeek Harness with ReMe's knowledge base to automatically capture main-agent conversations as user Markdown memory and schedule consolidation.

Who it is for — When managing main-agent conversations with DeepSeek Harness and wanting to capture completed dialogues as user-owned Markdown memory, this plugin provides integration. For users who do not want to share the workspace with ReMe, independent knowledge base tools offer a different perspective.

Watch out — The plugin requires manually allowing the build script when installing from source. Sandbox testing has not been completed. The plugin executes shell commands.

The verdict — If I am willing to manually allow the build script from source, I would install it because it connects conversations to persistent Markdown memory.

What it is — Integrates ReMe's local-first self-evolving personal knowledge base with DeepSeek Harness.

Who it is for — When running agents using DeepSeek Harness that require automatic capture of conversations as Markdown memory, this is suitable. If you do not need shared local workspace across agents, you can skip this plugin.

Watch out — No obvious pitfalls found. The plugin will execute shell commands.

The verdict — I would install it, but only if shell command execution is permitted.

What it is — MemSearch provides shared Markdown memory for DeepSeek Harness coding agents. It supports automatic capture, pre-step context injection, searchable recall, and memory-to-skill self-evolution through a review panel.

Who it is for — When building complex coding agents with DeepSeek Harness that require persistent context across sessions, MemSearch offers value. If your work involves only short-lived conversations without cross-platform memory needs, you can skip it.

Watch out — It will execute shell commands. Sandbox testing has not been done yet; no obvious issues were identified.

The verdict — I would install it because it enables memory to evolve into reusable skills, but only if you are willing to use it via subdirectory installation and accept shell command execution as a trade-off.

What it is — deja-vu indexes session files from other coding agents to provide memory recall for DeepSeek Harness.

Who it is for — When using DeepSeek Harness with agents like Claude Code for code debugging tasks referencing historical solutions, users missing cross-agent recall benefit from it. If your workflow is limited to a single local agent with no cross-agent session sharing, this plugin is irrelevant.

Watch out — Sandbox testing passed in a new profile with registration in DeepSeek Harness. Index build takes about ten seconds. No obvious issues found.

The verdict — I would install it because it uses local BM25 index with no LLM, no embeddings, and no network.

What it is — Indexes the sessions coding agents have written to disk for recall in Claude Code and others.

Who it is for — Users working in Claude Code who need to recall solutions from months ago. Users whose work does not involve multi-session coding workflows also do not need it.

Watch out — Requires manual approval of the build script when installing from source. Sandbox test passed. No obvious issues found.

The verdict — I would install it because it recalls existing session history from disk without needing additional model calls.

What it is — Graph Memory provides traceable, searchable cross-session memory for DeepSeek Harness.

Who it is for — If you are using DeepSeek Harness to perform multi-turn agent tasks requiring cross-session knowledge recall, this plugin is suitable. If you only complete tasks in a single session and do not need durable memory survival, you do not need to install it.

Watch out — Sandbox test shows installation succeeds but pnpm blocks the build script for @photostructure/sqlite@1.2.1, requiring manual allowance to complete registration. It requires providing an API key or token.

The verdict — I would install it because it provides native DSH integration and durable cross-session memory, but I must manually allow the build script first.

What it is — Provides BM25 + semantic RAG based failure recovery memory search and knowledge base for DeepSeek Harness agents.

Who it is for — When DeepSeek Harness agents are processing real engineering sessions and encounter repeated failure patterns, this plugin allows quick retrieval of historical lessons. For agents not using the MCP protocol, this plugin's tool interfaces remain unused.

Watch out — Sandbox test passed in a new profile where it was registered into the profile. No obvious issues found.

The verdict — I would install it because it supports remote MCP access without installation or account.

What it is — This plugin connects DeepSeek Harness agents to local Flowix notes via flowix-cli MCP, enabling search, read, create and edit of memos and mind maps.

Who it is for — It is suitable when software developers need coding agents to reference persistent project documents. If your work is limited to personal knowledge organization without agent integration, you may not need this plugin.

Watch out — Sandbox testing has not been performed. It requires the flowix-cli tool.

The verdict — I would install it if the agent needs to edit Flowix memos, but only if flowix-cli is ready.

What it is — Provides cross-agent persistent memory management for DeepSeek Harness powered by Mnemon.

Who it is for — When using multiple Mnemon-enabled agents that need shared long-term memory, this plugin applies. For users not running DeepSeek Harness, this plugin has no effect.

Watch out — Sandbox testing passed; it registers in the profile. Requires separate installation of Mnemon Native CLI.

The verdict — I would install it because it supports cross-agent long-term memory sharing, but only if Mnemon Native CLI is installed.

What it is — Adds cross-session long-term memory and skill management capabilities to DeepSeek Harness.

Who it is for — Suitable for developers managing multiple projects who need AI to remember project logs and todos. Not suitable for users who only develop on a single device because it provides memory synchronization.

Watch out — No obvious pits found, but it executes shell commands. Sandbox test passed. Installed into a new profile and auto registered.

The verdict — I would install it because it provides COI scheduling and session broadcast, suitable for developers wanting AI as a team commander.

What it is — Long-term memory for DeepSeek Harness stored as plain Markdown files.

Who it is for — It suits users who need to keep long-term memory in DeepSeek Harness. It does not suit users who hope the model automatically saves memories, because it requires joint decision by user and model.

Watch out — It requires an API key or token. It will execute shell commands. No obvious issues found.

The verdict — I would install it because it supports sharing memory storage with Claude Code, Codex, Kiro and OpenClaw, but only if the index can be compacted step by step.

What it is — This DeepSeek Harness plugin provides a deterministic capping protocol for the MEMORY.md index.

Who it is for — It is suitable when managing Markdown fact storage in DeepSeek Harness. Users performing writes via shell commands or external editors do not need it, since the index capping does not cover those paths.

Watch out — Installation requires providing an API key or token. Sandbox testing passed in a fresh profile and the plugin was registered. No obvious issues found.

The verdict — I would install it because it denies growth past the index cap for matched edit tools.

What it is — Provides project-scoped cross-session memory for DeepSeek Harness using SQLite to store seven layers of data and provide tools.

Who it is for — Users performing multi-turn project development tasks in DeepSeek Harness who lack structured cross-session memory management capabilities. Users handling only single-turn session memory may not need the idle window consolidation feature.

Watch out — Sandbox test passed, installed on new profile and registered into profile by harness. No obvious pitfalls found, though pnpm users may need to allow builds by editing pnpm-workspace.yaml during installation.

The verdict — If I were you, I would install it because it automates memory consolidation in idle windows, providing structured capability for multi-turn project work.

What it is — dsh-mneme is an AI memory engine plugin.

Who it is for — When using DeepSeek Harness, if the agent needs to simulate real brain memory mechanisms, it is suitable to install it. If the user does not use the JavaScript language, they do not need to install it.

Watch out — Sandbox test passed: installed in a new profile and registered into the profile by harness. Static check indicates it will execute shell commands. No obvious pitfalls found.

The verdict — I will not install it because it executes shell commands.

What it is — dsh-mneme is a structured memory engine for DeepSeek Harness.

Who it is for — It is suitable in scenarios where DeepSeek Harness projects require maintaining context across sessions. Users who do not want the plugin to execute shell commands can skip installing it.

Watch out — Static check indicates that the plugin executes shell commands. Sandbox test passed on a new profile, where it was registered in the harness profile.

The verdict — I would install it because it supports entity-attribute-timeline and human-editable Markdown storage.

What it is — Integrates Noema to deliver durable inspectable long-term memory for DSH agents, including tools for recall search browse knowledge-graph and import from other AI coding tools.

Who it is for — When using DSH agents for coding tasks requiring knowledge retention across sessions, it fills the gap in durable memory. If your work does not involve importing memories from other AI coding tools, you can skip it.

Watch out — It executes shell commands at runtime to maintain the memory server. Sandbox testing passed with successful registration in a fresh profile. No obvious issues found.

The verdict — I would install it because it supports hot reload without needing a restart after the first boot, but the premise is using DSH 0.1.1-rc.1 or later.

What it is — dsh-mneme is a DeepSeek Harness plugin that provides cross-session memory with automatic consolidation, conflict freezing, and local semantic search.

Who it is for — If you are conducting multi-turn continuous conversations in DeepSeek Harness and need the AI to recall user preferences and prior project details, then install this plugin. If your work is limited to single interactions without cross-session memory management, then no installation is needed.

Watch out — Sandbox testing passed in a new profile with registration in harness. Static checks indicate shell command execution but no other issues found.

The verdict — I would install it because it supports agent-scoped memory isolation and editable Markdown mirrors, which add value in complex multi-agent setups.

What it is — Provides approval-gated auditable cross-session memory via typed ctx.memory with SQLite provider and frozen snapshot for DeepSeek Harness.

Who it is for — If you need to maintain user and agent memory when using any model, it is suitable to install. If you only work within a single session, you can skip it.

Watch out — Sandbox test passed, installed in a new profile and registered to harness. No obvious issues found.

The verdict — I would install it because it provides an approval gate the model cannot bypass for writes.

What it is — Provide cross-session SQLite memory storage and tool support for DeepSeek Harness.

Who it is for — If you are doing multi-turn project discussions in DeepSeek Harness requiring maintenance of user and project level memory, you can use this plugin. Users who only perform single-turn tasks may find it unnecessary.

Watch out — Tested successfully in a fresh profile where it registers with Harness. If using pnpm, manual build allowance is needed. No obvious pitfalls found.

The verdict — I would install it because it offers a self-contained solution with per-window consolidation for long sessions.

What it is — dsh-mneme is a cross-session memory plugin that recalls across sessions, auto-consolidates memories, freezes conflicting memories for review, and keeps a replayable audit trail with local semantic search over SQLite.

Who it is for — When using DeepSeek Harness for multi-agent projects that require cross-session agent isolation and preference retention with local semantic search. If operating only within a single session without needing isolation or audit features.

Watch out — Sandbox test passed in a fresh profile with successful registration. Static check indicates execution of shell commands. No obvious pitfalls found.

The verdict — I would install it because of the cross-session memory and agent isolation, but only if shell command execution is allowed in the environment and memory deletion is not required.

What it is — It provides model-driven context compression for DeepSeek Harness, letting the model decide when and what to compress.

Who it is for — When using DeepSeek Harness to handle long conversations where context is exhausted and you need to retain key details from early messages, this plugin suits you. Users who do not want the model to decide compression can continue using DSH's built-in automatic compression.

Watch out — As a test version plugin, please do not use it in production or engineering environments. Sandbox test passed: it was installed in a fresh profile and registered into the Harness profile. No obvious pits found.

The verdict — I would install it because it turns compression ranges into recoverable checkpoints where key conclusions keep accumulating.