Documentation
¶
Overview ¶
@index Suspect fallback call-edge analysis based on overlapping annotation vocabulary between source and target nodes.
Index ¶
Constants ¶
This section is empty.
Variables ¶
This section is empty.
Functions ¶
This section is empty.
Types ¶
type Result ¶
type Result struct {
Items []SuspectEdge
Pagination paging.Page
}
Result carries one suspect fallback edge page plus pagination metadata. @intent let callers expose bounded fallback suspect responses while preserving legacy fields.
type Service ¶
type Service struct {
// contains filtered or unexported fields
}
@intent bundle graph DB and annotation-store access needed to detect low-confidence fallback edges.
func New ¶
func New(db *gorm.DB, graphStore store.GraphStore) *Service
@intent construct the suspect fallback analyzer from the active graph DB and annotation-capable store.
func (*Service) FindSuspects ¶
@intent report fallback call edges whose source and target annotations share no intent or domain-rule vocabulary. @param ctx carries the namespace that ctxns.FromContext uses to scope analysis. @param opts carries pagination bounds for the returned suspect edge window. @return returns suspect edges sorted by source qualified name and then target qualified name. @domainRule exclude an edge from suspect results as soon as source and target share any intent/domainRule token. @see mcp.handlers.findSuspectFallbackEdges
func (*Service) FindSuspectsPage ¶
FindSuspectsPage reports suspect fallback call edges and returns one bounded page. @intent bound fallback suspect analysis before annotation lookups so large graphs cannot expand unbounded responses. @domainRule fetch limit+offset+1 fallback edges before suspect filtering; paging bounds the scanned fallback-edge window, so returned suspect items may be fewer than limit while HasMore still reflects additional fallback edges beyond that window.
Source Files