Codebase Memory — Structural Graph

Zero-file-read structural analysis. Always try these tools BEFORE reading files for structural questions.

Quick Decision Matrix

Question	Tool
What calls foo()?	trace_path(function_name="foo", direction="inbound")
What does foo() call?	trace_path(function_name="foo", direction="outbound")
Full call chain	trace_path(direction="both", depth=3)
Impact of my change	detect_changes(base_branch="main")
Architecture overview	get_architecture
Find by name/pattern	search_graph(name_pattern="Auth.*")
Find by label	search_graph(label="Class")
Dead code	search_graph(max_degree=0, label="Function")
Custom query	query_graph(query="MATCH (n:Function)-[:CALLS]->(m) RETURN n.name, m.name LIMIT 20")
Execution flows	list_flows → get_affected_flows
ADR management	manage_adr(mode="get")

Workflow — First Use

1. index_repository(repo_path=".") — first run only (~6s for 500 files)
2. list_projects — verify indexed
3. get_architecture — architecture map

Workflow — Before Reviewing Changes

1. detect_changes(base_branch="main") → risk score + blast radius
2. get_affected_flows → which execution paths break
3. trace_path on highest-risk functions

Workflow — Exploration

1. get_architecture → community map, entry points
2. search_graph(name_pattern="<keyword>") → find relevant nodes
3. trace_path → traverse from found nodes
4. get_code_snippet(qualified_name="<name>") → source with context

Known Gotchas

1. search_graph(relationship="HTTP_CALLS") filters by degree — use query_graph for actual edge inspection
2. query_graph has a 200-row cap — use search_graph with degree filters for counts
3. trace_path requires exact qualified names — use search_graph(name_pattern=...) first
4. direction="outbound" misses cross-service callers — use direction="both"
5. Results default to 10 per page — check has_more and paginate with offset

Re-indexing

Graph auto-updates via background git polling (5–60s adaptive). For immediate refresh: index_repository(repo_path=".", mode="fast")