hivemind-graph

1# Hivemind Code Graph
2 
3A deterministic, AST-derived map of the current repository — every function,
4class, method, interface, type, enum, const, and module, plus the edges between
5them (`calls`, `imports`, `extends`, `implements`, `method_of`). It is queried as
6synthesized files under the Deeplake mount; there are no real files on disk and
7no network call in the read path.
8 
9The graph **builds and refreshes automatically** (on Stop / SessionEnd, gated by
10a rate limit + git diff). You never run a build command — just read it.
11 
12Use it as a fast **INDEX** to locate the few files/symbols that matter, then open
13them with `Read` to answer. It is not a substitute for the source.
14 
15## When to use this skill
16 
17Activate when the user asks a *structural / relational* question about the code:
18 
19- "What calls `pushSnapshot`?" / "Who uses this function?"
20- "What does `deeplake-pull.ts` import?" / "What depends on X?"
21- "Where is `GraphSnapshot` defined?" / "Find the function that handles Y."
22- "What are the main subsystems / the architecture here?"
23- "If I change this signature, what's affected?" → use `impact/<symbol>` (transitive blast radius)
24 
25## When NOT to use this skill
26 
27- Reading the **body** of a symbol you already located → use `Read` on the real
28  source file. The graph gives location + relationships, not full source.
29- Code that isn't **committed/built** yet — the graph can lag uncommitted edits.
30  If a file's mtime is newer than the build timestamp, read the live source.
31…