$ loading_
Add local semantic code search and persistent memory to AI code editors.
Copy the install command and let the AI configure it · recommended for beginners
Please install the "Vectr" MCP server from askskill: Run: claude mcp add 'io-github-swapnanil-vectr' -- uvx vectr
Use Vectr to run a semantic search across this repository. Find code, configuration, and tests related to user login and session management, then summarize how they connect.
Relevant files or snippets, plus a structured explanation of the login and session flow.
Save the current refactoring task’s goals, completed work, remaining tasks, and risks into Vectr’s working memory so we can continue later.
Persistent task context stored for reuse in later editing sessions.
Using Vectr’s local code search, identify possibly duplicated data validation logic in this repository and list modules that could be consolidated.
A list of suspected duplicate implementations and suggestions for consolidation.
Developers taking over an unfamiliar project can use it for semantic code search to locate implementations, configs, and tests tied to a feature. This makes it easier to build a high-level understanding than with keyword search alone.
During multi-step coding or refactoring work, it can preserve persistent working memory for an AI code editor. When users return later, they do not need to restate the task background and progress from scratch.
For teams or individuals who do not want to rely on external API keys, it suits local environments by adding search and memory capabilities to AI editors. This reduces setup requirements tied to external services.
It is an MCP tool that provides semantic codebase search and persistent working memory for AI code editors. The description also says it runs locally and does not require an API key.
Based on the provided description, no API key is required. It is described as a local tool.
The provided information indicates that it supports semantic codebase search rather than simple keyword matching, and it also adds persistent working memory. For more detailed differences, see the source repository.
Give AI coding assistants local persistent memory, semantic search, and task tracking.
Let AI coding agents search local code by meaning, not exact keywords.
Search indexed codebases semantically with natural language across MCP clients.
Search codebases semantically with natural language to find relevant files and logic.
Search local codebases semantically with natural language and precise retrieval.
Search and navigate multiple code repositories with natural language understanding.