Understand chat
Skill ComeOnOliver/skillshub/skills/Lum1104/Understand-Anything/understand-chat
π§ The right skill, one API call. AI agent skills registry with token-efficient skill resolution. 5,000+ skills from 500+ top repos.From the repository description
npx -y skills add ComeOnOliver/skillshub --skill understand-chatAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
SKILL.md
2.5 KB, 600 tokens by cl100k_base, as published. Nobody here has run it
/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:
projectβ {name, description, languages, frameworks, analyzedAt, gitCommitHash}nodes[]β each has {id, type, name, filePath, summary, tags[], complexity, languageNotes?}- Node types: file, function, class, module, concept
- IDs:
file:path,func:path:name,class:path:name
edges[]β each has {source, target, type, direction, weight}- Key types: imports, contains, calls, depends_on
layers[]β each has {id, name, description, nodeIds[]}tour[]β each has {order, title, description, nodeIds[]}
How to Read Efficiently
- Use Grep to search within the JSON for relevant entries BEFORE reading the full file
- Only read sections you need β don't dump the entire graph into context
- Node names and summaries are the most useful fields for understanding
- Edges tell you how components connect β follow imports and calls for dependency chains
Instructions
-
Check that
.understand-anything/knowledge-graph.jsonexists in the current project root. If not, tell the user to run/understandfirst. -
Read project metadata only β use Grep or Read with a line limit to extract just the
"project"section from the top of the file for context (name, description, languages, frameworks). -
Search for relevant nodes β use Grep to search the knowledge graph file for the user's query keywords: "$ARGUMENTS"
- Search
"name"fields:grep -i "query_keyword"in the graph file - Search
"summary"fields for semantic matches - Search
"tags"arrays for topic matches - Note the
idvalues of all matching nodes
- Search
-
Find connected edges β for each matched node ID, Grep for that ID in the
edgessection to find:- What it imports or depends on (downstream)
- What calls or imports it (upstream)
- This gives you the 1-hop subgraph around the query
-
Read layer context β Grep for
"layers"to understand which architectural layers the matched nodes belong to. -
Answer the query using only the relevant subgraph:
- Reference specific files, functions, and relationships from the graph
- Explain which layer(s) are relevant and why
- Be concise but thorough β link concepts to actual code locations
- If the query doesn't match any nodes, say so and suggest related terms from the graph
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.