Gabeujin Service Endpoint Tracer
Use this skill when a project needs evidence-backed tracing from UI calls or service identifiers into backend endpoints and storage layers.
Good Fits
- Legacy enterprise applications with controller -> service -> mapper layering
- Frontend service IDs or URL calls that need a backend owner
- Mapper or SQL IDs that must be traced back to an endpoint
- Migration and impact analysis work where hidden dependencies matter
Trace Procedure
- Start from the strongest known identifier:
- service ID
- URL or route
- controller method
- service method
- mapper or SQL ID
- Follow the call chain one hop at a time and record evidence at each hop.
- Distinguish confirmed links from inferred links.
- Capture side inputs such as common code lookups, shared utilities, or message resources when they affect the flow.
- End with a compact path summary that names the files, methods, and boundaries involved.
Output Format
- Entry point
- Trace path
- Supporting evidence
- Impacted layers
- Open questions
- Verification suggestions
Rules
- Do not invent links that are not supported by repository evidence.
- Mark inferred mappings explicitly.
- Prefer tight, file-backed traces over broad speculation.
- When a trace crosses multiple subsystems, note the blast radius.