EchoVault FAQs

Question 1

What is EchoVault and how does it help coding agents?

Accepted Answer

EchoVault provides local, persistent memory for AI coding agents (like Claude Code, Cursor, Codex, OpenCode). It allows agents to store decisions, bugs, and context from their sessions as Markdown files, enabling efficient retrieval and continuity across tasks without external dependencies or cloud costs.

Question 2

Does EchoVault support both keyword and semantic search?

Accepted Answer

Yes, EchoVault features a hybrid search system. It provides fast FTS5 keyword search out-of-the-box. For enhanced understanding, you can optionally configure semantic vector search using local models via Ollama or cloud providers like OpenAI, offering more relevant context retrieval.

Question 3

How are memories stored and what are the benefits of this approach?

Accepted Answer

Memories are stored locally as Obsidian-compatible Markdown files, one per session per project, within a designated vault. This local-first approach ensures privacy, zero idle cost, and allows you to browse and manage your agent's memories visually using Obsidian or any text editor.

Question 4

Is EchoVault compatible with multiple AI coding agents?

Accepted Answer

Absolutely. EchoVault is designed to be cross-agent compatible, working with Claude Code, Cursor, Codex, and OpenCode. A memory saved by one agent is fully searchable and accessible by any other supported agent, ensuring a unified memory vault across your development workflow.

Question 5

How does EchoVault handle sensitive information?

Accepted Answer

EchoVault incorporates 3-layer secret redaction to protect sensitive data like API keys and passwords. It supports explicit `<redacted>` tags, pattern detection for common secrets, and custom `.memoryignore` rules, ensuring confidential information is stripped before storage.

EchoVault

EchoVault

Key Features

Use Cases

Key Features

Use Cases