$ loading_
Give Codex agents encrypted local memory across sessions for project context.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "Perseus Vault Codex" yet — see the docs or source repo.
Save these project conventions to persistent memory: use TypeScript strict mode for backend code; run tests before commits; return standardized JSON for API errors.
The tool stores the coding conventions in persistent memory for future Codex sessions.
Remember this decision: switch authentication to JWT for stateless deployment; keep the user profile service separate from the main app.
The tool saves key technical decisions and rationale for reuse as future context.
Read the previously saved memory for this project and summarize conventions, decisions, and context that should be continued.
Returns a summary of prior project memory so the agent can quickly resume work in a new session.
Developers working with Codex on long-running projects can store naming rules, commit requirements, and API conventions in persistent memory. New sessions can then continue following the same rules.
When a project goes through architecture changes or tradeoff decisions, this tool helps the agent retain key choices and their rationale. Later sessions can continue without re-explaining the context.
For teams that prefer local-first storage and care about memory security, this tool provides encrypted persistent local memory. It fits development workflows that need less context loss.
It is a small MCP server that wraps Perseus Vault. It gives OpenAI Codex agents persistent, encrypted, local-first memory for project conventions, decisions, and context.
Based on the description, it is mainly for users who work with OpenAI Codex agents, especially developer teams that need project context to continue across sessions.
The provided information does not include installation steps, runtime requirements, or key configuration details. Please see the source repository for exact setup and prerequisites.
Store and search encrypted local memory for AI agents efficiently.
Provide durable local memory for Codex and other MCP-enabled agents.
Persist, search, and retrieve coding knowledge and session notes for agents.
Persistent memory for coding agents across sessions
Provide local-first memory and retrieval for coding agents using SQLite embeddings.
Give AI coding agents persistent semantic memory with embedding-based retrieval.