Free tools Windows power users keep installed
One-click scans. No signup required.
AutoMemoryTools gives a Spring AI agent a file-based way to carry selected facts from one conversation into later ones. It is designed for curated information—such as a user preference or a project decision—not as a complete transcript. You can register its memory tools and companion system prompt in a ChatClient, or use the project’s advisor-based integration.
What AutoMemoryTools remembers
AutoMemoryTools stores selected information in Markdown files under a configured memories directory. Those files can persist across application runs, giving later conversations a way to consult useful facts without treating every earlier message as long-term memory. The project documents six operations for viewing, creating, editing, inserting, deleting, and renaming memory files; operations are scoped to the configured root. See the AutoMemoryTools documentation for the current behavior and setup details.
As an Amazon Associate I earn from qualifying purchases.
This is a curation model: the application and agent work with memory entries chosen to be useful later, rather than relying on an entire conversation log. The project’s demo illustrates saving a person’s name, role and response preference, along with a project migration decision, then asking about them in a separate run. That is an example of the documented pattern, not a guarantee that every model will recall or apply every fact correctly.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
How the memory files and index fit together
Typed Markdown entries
Each memory entry is a Markdown file with YAML frontmatter. The documented types include user, feedback, project and reference. Frontmatter also provides a short name and description, helping identify an entry’s subject and purpose.
#1 Best Overall
The MEMORY.md index
A MEMORY.md file acts as an index pointing to individual entries. It provides an always-loaded list and guidance for selecting relevant memories. This separates the compact overview from the details in individual files: an agent can use the index to identify which entries matter to a new request rather than loading a full transcript as memory.
Wire it into a Spring AI ChatClient
The project documents two integration shapes: register AutoMemoryTools and its companion system prompt directly in the ChatClient setup, or use the AutoMemoryTools advisor described in the project’s Spring article. The demo shows the manual pattern: configure a persistent memory directory, supply the prompt template and default tools, and include a tool-call advisor. Consult the Memory Tools Demo and the project documentation for current dependency coordinates, provider configuration, APIs and example code; those details can change.
- Choose a persistent root. Configure the memories directory where the files and index will live. The demo uses a directory intended to survive process restarts; an ephemeral location will not provide that persistence.
- Provide the companion prompt. The prompt is part of the documented setup and guides the agent’s use of the memory files and tools. Registering tools alone is not the complete documented wiring pattern.
- Register the tools and tool-call handling. Add
AutoMemoryToolsto the ChatClient’s configured tools and use the tool-call advisor as shown in the demo. Alternatively, follow the project’s advisor-based integration where that better fits the application. - Configure the provider. The demo requires an AI provider configuration. Use the provider and model settings appropriate to your application rather than copying potentially stale identifiers from an older example.
- Try a cross-session example. The project demo uses the question “What do you know about me?” after saving illustrative facts in an earlier run. Treat the result as a demonstration of the intended flow, not a benchmark or a promise of model behavior.
AutoMemoryTools versus Spring AI ChatMemory
Spring AI ChatMemory is a separate abstraction for storing and retrieving conversation messages through a ChatMemoryRepository. It addresses message history, while AutoMemoryTools addresses curated facts in files. These patterns can complement one another; neither should be assumed to replace every need addressed by the other.
Recommended Free Tools
| Question | AutoMemoryTools | Spring AI ChatMemory |
|---|---|---|
| What is retained? | Selected, curated facts and references in memory files. | Conversation messages managed through a repository. |
| Where is it stored? | Markdown files under a configured memories root. | A ChatMemoryRepository; documented implementations include in-memory and persistent options. |
| How is information selected? | A MEMORY.md index points to entries, with guidance for selecting relevant memories. |
Message storage and retrieval follow the repository and application’s chat-memory configuration. |
| What about tool-call messages? | The project documents memory-file operations; this is not a transcript repository. | The current JDBC reference says assistant messages containing tool calls and tool response messages are filtered when saved. |
| What operational setup fits? | A configured file directory and the AutoMemoryTools prompt/tool integration. | A chosen repository implementation, which may be in-memory or backed by a database. |
The Spring AI Chat Memory reference lists JDBC, Cassandra, Neo4j, MongoDB and Redis among repository options. These are choices for chat-message storage, not interchangeable backends for AutoMemoryTools’ Markdown-file model. If your requirement is retaining conversation history, choose and configure ChatMemory accordingly; if it is carrying a small set of useful facts between sessions, the file-based pattern addresses that different need.
Security and design considerations
The project documentation says memory operations are scoped to a sandboxed memories root and that path traversal and absolute-path injection are blocked. This is a statement about the project’s documented protections, not an independent security audit. Applications should still control access to the configured directory and evaluate it within their own threat model.
- Decide what deserves long-term retention. Curated files are not a reason to save every message. Keep entries focused on information that will genuinely help future sessions.
- Plan for edits and deletion. The documented operations include editing and deleting entries, so define who or what may make those changes and how the files fit the application’s retention practices.
- Choose the storage pattern for the job. File-based facts do not supply the same function as persisted message history. Consider retention controls, operational fit and tool-call-message needs before choosing one or both approaches.
Origins and scope
The project describes AutoMemoryTools as inspired by Claude Code memory conventions and Anthropic’s Memory Tool specification. Its documentation says each AutoMemoryTools method maps one-to-one to an operation in that specification. Spring’s Spring AI Agentic Patterns, Part 6, by Christian Tzolov and published April 7, 2026, presents it as a Spring AI port of these memory patterns. These are descriptions of the project’s design and lineage, not independent findings about comparative performance or effectiveness.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
