$ loading_
Build a queryable code graph, validate edit scope, and log reasoning.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "agent-context-graph" yet — see the docs or source repo.
Scan the current project, build a file and symbol knowledge graph, and find which files and functions call UserService.login; before I modify it, list the impacted scope and potential risks.
A call graph for the login method, related files, affected modules, and pre-edit risk notes.
Only allow edits to code directly related to the login flow in src/auth/ and src/api/auth.ts. If changes outside this scope are needed, stop first, explain why, and wait for my approval.
The agent edits only within the allowed scope; if out-of-scope changes are required, it explains why and suggests next steps without writing files.
Output the append-only reasoning log for this code change, showing in time order which symbols you queried, why you changed these files, and what each edit was meant to fix.
An auditable change log that clearly records the query process, decision rationale, and purpose of each modification.
Give AI coding agents filesystem, Git, database, and compute tools via MCP.
Enable AI coding agents to communicate, share state, and coordinate work in real time.
Give AI coding agents persistent code memory while cutting token usage dramatically.
Query code structure and cross-language relationships via MCP with auditable access logs.
Give AI coding agents persistent local memory across sessions.
Build a knowledge graph from repos for Q&A and implementation planning.