Add project instructions with rules
JetBrains project rules can describe your code, preferred tools, and coding conventions. They are configured under Settings | Tools | AI Assistant | Rules.
Create a rules file
Creating a project rules file creates an .aiassistant/rules folder with a Markdown file. You can also create .aiassistant/rules in the project root and add a Markdown file manually. See JetBrains AI Assistant project rules.
Choose when rules apply
Each rules file has a selectable rule type. Always rules apply to all chat sessions. Manually applied rules can be invoked with @rule: or #rule:, or attached with Add attachment. See JetBrains AI Assistant project rules.
Use model and file-based rules
By model decision rules apply when the model considers them relevant, so include an Instruction explaining when to apply them. By file patterns rules apply when a chat-referenced file matches a pattern such as *.kt or src/**. See JetBrains AI Assistant project rules.
Write useful guidelines
Put your guidelines in the rules Markdown file. JetBrains examples cover naming, style, structure, best practices, documentation, and tools. You can expand the attachments list at the start of a response to check whether rules were applied. See JetBrains AI Assistant project rules.
Connect KnowMan for saved knowledge
KnowMan MCP is a local stdio server started with npx -y @knowman/mcp and requires Node 22 or newer. It uses a KNOWMAN_API_KEY Bearer key, or a saved CLI login. See the MCP server guide.
Where KnowMan fits
KnowMan can search saved items, read one item, answer questions with citations, and save URLs. The tools are kb_search, kb_get_item, kb_ask, and kb_save. KnowMan runs locally with npx -y @knowman/mcp and as a hosted server at https://api.knowman.io/mcp with OAuth sign-in or an API key.
Use the hosted KnowMan server
In clients that support remote MCP servers, add https://api.knowman.io/mcp, then sign in with OAuth on the KnowMan consent page or send an API key as an Authorization: Bearer <key> header. For example, Cursor mcp.json can use {"mcpServers":{"knowman":{"url":"https://api.knowman.io/mcp","headers":{"Authorization":"Bearer <key>"}}}}.
Then ask the assistant to search your KnowMan for saved knowledge or save a URL. The hosted server exposes the same kb_search, kb_get_item, kb_ask, and kb_save tools. The open-source local client is available at https://github.com/Essatt/knowman-mcp.
Questions people ask
How do I give JetBrains AI Assistant project rules?
Open Settings | Tools | AI Assistant | Rules, then create or edit a Markdown file in .aiassistant/rules. Choose a rule type and add your guidelines. See JetBrains AI Assistant project rules.
Can JetBrains AI Assistant use saved articles as knowledge?
JetBrains project rules are documented for project-specific guidance, but a general feature for adding saved articles or reference documents is not documented here. A KnowMan MCP connection may provide access to saved items if the client supports a local stdio MCP server.
Can KnowMan teach JetBrains AI Assistant my knowledge?
KnowMan MCP exposes tools to search saved items, read one item, ask questions with citations, and save URLs. It runs locally with npx -y @knowman/mcp and as a hosted server at https://api.knowman.io/mcp with OAuth sign-in or an API key.
What are the KnowMan MCP tools?
kb_search searches saved items, kb_get_item reads one item, kb_ask answers with citations, and kb_save saves a URL. KnowMan runs locally with npx -y @knowman/mcp and as a hosted server at https://api.knowman.io/mcp with OAuth sign-in or an API key.