Connect code structure to the decisions around it.
A repository explains what the software does; saved decisions explain why. Index selected checkouts so your coding assistant can investigate both from its connected workflow.
Illustrative workflow
Where is this behavior implemented, and what decision introduced it?
Query the indexed definitions and callers, retrieve related project notes, then inspect the current repository files before proposing a change.
Illustrative scenario, not a live product result.Select the checkouts to index.
The operator configures local repository paths through MEMRAIN_CODE_PATHS and makes those checkouts available to the deployment. Indexing reads the selected source files; managing repository credentials, updating the checkout and choosing its branch remain part of that environment. Keep those choices explicit so the assistant knows which code state the index represents.
Build a navigable code view.
Memrain extracts definitions, references and call relationships for supported TypeScript, Python and Go code. A connected client can use that structure to find callers, follow a request path or investigate the likely impact of a change. Source files are screened for credentials before code indexing by default. Relevant decision pages and imported coding sessions can add the surrounding history.
Check freshness and coverage.
An indexed checkout is evidence about the code it contains. Verify its revision and the graph's readiness before treating a result as current. Unsupported syntax, parse failures or an unfinished index can limit coverage, and an empty result from an unready graph is inconclusive. Use the retrieved paths to focus repository inspection and tests when making the actual change.
Keep the context. Move the work forward.
Connect your knowledge to the AI you already use. Find your starting point and make the next conversation count.