lum1104/understand-anything
8 skills10.6k installs620k starsGitHub
Install
npx skills add https://github.com/lum1104/understand-anythingSkills in this repo
1UnderstandThe understand skill Analyze a codebase to produce an interactive knowledge graph for understanding architecture, components, and relationships It covers $ARGUMENTS may contain:. Key workflows include --full - Force a full rebuild, ignoring any existing graph. Developers invoke understand when the task matches the triggers and reference files in SKILL.md for grounded, stepwise execution. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation. Reference files and progressive disclosure keep context focused while preserving concrete commands, configuration fields, and validation checks copied from the upstream documentation.2kinstalls2Understand DashboardThe understand-dashboard skill is designed for launch the interactive web dashboard to visualize a codebase's knowledge graph. /understand-dashboard Start the Understand Anything dashboard to visualize the knowledge graph for the current project. Determine the project directory: If $ARGUMENTS contains a path, use that as the project directory Otherwise, use the current working directory 2. Invoke when the user asks about understand dashboard or related SKILL.md workflows.1.4kinstalls3Understand ExplainThe understand-explain skill use when you need a deep-dive explanation of a specific file, function, or module in the codebase # /understand-explain Provide a thorough, in-depth explanation of a specific code component. ## Graph Structure Reference The knowledge graph JSON has this structure: - `project` - {name, description, languages, frameworks, analyzedAt, gitCommitHash} - `nodes[]` - each has {id, type, name, filePath?, summary, tags[], complexity, languageNotes?} - Code node types: file, function, class, module, concept - Non-code node types: config, document, service, table, endpoint, pipeline, schema, resource - Domain/knowledge node types: domain, flow, step, article, entity, topic, claim, source - IDs use the node type as prefix, e.g. `file:path`, `function:path:name`, `config:path`, `article:path` - `edges[]` - each has {source, target, type, direction, weight} - Key types: imports, contains, calls, depends_on, configures, documents, deploys, triggers, contains_flow, flow_step, related, cites - `layers[]` - each has {id, name, description, nodeIds[]} - `tour[]` - each has {order, title, description, nodeIds[]} ## How to Read Efficiently 1.1.3kinstalls4Understand ChatThe understand-chat skill use when you need to ask questions about a codebase or understand code using a knowledge graph # /understand-chat Answer questions about this codebase using the knowledge graph at `.understand-anything/knowledge-graph.json`. ## Graph Structure Reference The knowledge graph JSON has this structure: - `project` - {name, description, languages, frameworks, analyzedAt, gitCommitHash} - `nodes[]` - each has {id, type, name, filePath?, summary, tags[], complexity, languageNotes?} - Code node types: file, function, class, module, concept - Non-code node types: config, document, service, table, endpoint, pipeline, schema, resource - Domain/knowledge node types: domain, flow, step, article, entity, topic, claim, source - IDs use the node type as prefix, e.g. `file:path`, `function:path:name`, `config:path`, `article:path` - `edges[]` - each has {source, target, type, direction, weight} - Key types: imports, contains, calls, depends_on, configures, documents, deploys, triggers, contains_flow, flow_step, related, cites - `layers[]` - each has {id, name, description, nodeIds[]} - `tour[]` - each has {order, title, description, nodeIds[]} ## How to Read Effi.1.3kinstalls5Understand OnboardThe understand-onboard skill use when you need to generate an onboarding guide for new team members joining a project # /understand-onboard Generate a comprehensive onboarding guide from the project's knowledge graph. ## Graph Structure Reference The knowledge graph JSON has this structure: - `project` - {name, description, languages, frameworks, analyzedAt, gitCommitHash} - `nodes[]` - each has {id, type, name, filePath?, summary, tags[], complexity, languageNotes?} - Code node types: file, function, class, module, concept - Non-code node types: config, document, service, table, endpoint, pipeline, schema, resource - Domain/knowledge node types: domain, flow, step, article, entity, topic, claim, source - IDs use the node type as prefix, e.g. `file:path`, `function:path:name`, `config:path`, `article:path` - `edges[]` - each has {source, target, type, direction, weight} - Key types: imports, contains, calls, depends_on, configures, documents, deploys, triggers, contains_flow, flow_step, related, cites - `layers[]` - each has {id, name, description, nodeIds[]} - `tour[]` - each has {order, title, description, nodeIds[]} ## How to Read Efficiently 1.1.3kinstalls6Understand Diffunderstand-diff compares current git changes or pull requests to the living knowledge graph stored at `.understand-anything/knowledge-graph.json`. The graph records project metadata, languages, frameworks, and nodes for files, functions, classes, modules, configs, services, tables, endpoints, pipelines, and schemas—each with summaries, tags, and complexity scores. Developers reach for understand-diff when a diff is too large to mentally trace or when reviewers must see which components, services, and endpoints a PR touches. The skill translates raw patches into structured impact context so agents explain what changed, what depends on it, and where regression risk concentrates.1.2kinstalls7Understand Domainunderstand-domain is a skill from lum1104/understand-anything that scans any codebase and produces clean, structured domain context for LLM agents. The bundled extract-domain-context.py script walks a project root with limits of depth 6, 50 files per directory, 5000 files total, 40 sampled files, and capped lines per file, writing output to .understand-anything/intermediate/domain-context.json. A domain-analyzer agent then identifies business domains, flows, and steps. Developers reach for understand-domain when agents need accurate implementation or analysis in unfamiliar repositories without manual architecture tours.1.1kinstalls8Understand Knowledgeunderstand-knowledge is part of the understand-anything project and runs merge-knowledge-graph.py to combine a deterministic scan-manifest.json with LLM analysis batches (analysis-batch-*.json) into a final assembled knowledge graph. The script handles entity deduplication, edge normalization, layer building from index.md categories, and tour generation from index.md section ordering. Output is written to assembled-graph.json under .understand-anything/intermediate/ inside the wiki directory. Developers reach for understand-knowledge after codebase scans and LLM analysis batches exist and they need one merged, queryable graph instead of fragmented JSON artifacts.1.1kinstalls