What problem does Ummon solve?
Ummon helps developers and AI assistants better understand complex codebases by creating a searchable, interconnected knowledge graph of code entities and domain concepts.
@Nayshins
基本信息
The semantic layer for software engineering: Connect code to meaning, build on understanding
工具列表暂未提供
当前收录暂时没有可展示的工具列表,不代表服务没有工具。请查阅项目文档了解支持的能力。
工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。
Ummon is a code analysis tool that builds knowledge graphs from codebases to enhance understanding, improve AI assistance, and enable sophisticated querying. It creates connections between code entities (functions, classes, modules) and domain concepts, making it easier to reason about complex software systems and identify relevant code for specific tasks. The project is in early development and is not yet stable.
Install Ummon with cargo install ummon, then use the CLI to index a codebase (ummon index /path/to/codebase), query it (ummon query "show all authentication functions"), or get AI-assisted recommendations (ummon assist "implement a user registration function"). Set the OPENROUTER_API_KEY environment variable for LLM features.
Ummon helps developers and AI assistants better understand complex codebases by creating a searchable, interconnected knowledge graph of code entities and domain concepts.
Ummon supports Rust, Python, JavaScript, and Java. Each language parser extracts entities like classes, functions, and imports.
By default, Ummon performs incremental updates — it detects modified files and reindexes only those. Use `--full` for a complete rebuild of the knowledge graph.
Ummon requires the `OPENROUTER_API_KEY` environment variable for LLM-powered features (queries and domain extraction). All other configuration is via command-line flags.
No. The README explicitly warns that the project is in early development and is not yet stable. APIs and functionality may change significantly between versions.
OrangePro local-first CLI + MCP server for behavior mapping, grounded test generation, and dynamic proof.

Local-first credential broker for MCP. AI coding agents (Claude Code, Cursor, Codex…) call APIs, SSH and databases through scoped presets; secrets stay in your OS keychain and never enter the agent's context.
Modelglass turns "which model should I use?" into a lookup instead of a guess. Seven tools give your agent live pricing across image, language, video, and audio models. Sourced and versioned, not scraped once and forgott
Get a second opinion on your code from every coding agent you have. Claude Code, Codex and Grok review each other's diffs and answer each other's questions, even inside a session that's already running. Uses the CLIs you

Catch intent conflicts before code conflicts. The open-source coordination protocol for coding agents, built above Git.
评论