// Systematically debug issues using the code-review-graph knowledge graph. Use instead of manual file reading when tracing call chains, locating a bug's origin, or assessing impact radius of a suspected file.
Use ADB to connect devices, install Android debug builds, and troubleshoot deployment failures. Use for device detection errors, install failures, launch failures, package selection across flavors, and first-pass process checks. For deeper runtime investigation, prefer the Argent workflow.
Investigate Android runtime bugs with evidence-first Argent workflows. Use for emulator targeting, runtime console/network inspection, UI hierarchy capture, optional native traffic inspection, and root-cause analysis before changing code.
Capture quick Android UI evidence with explicit launches, UIAutomator dumps, and adb screenshots for fast visual debugging and reproducible repro notes.
Instruction-to-skill navigation map. Use when routing tasks to the right instruction and skill files or migrating legacy guidance to the split-doc model.
Step-by-step module creation checklist. Use when adding a new Gradle module and wiring build, DI, and architecture integration correctly.
Android platform/helper layer guide for AniTrend. Use when working in `:android:*` modules or deciding whether to reuse or extend existing Android-side helpers for configuration/theme, context or fragment utilities, notification permission flows, deep links, drawer/app-shell behavior, or other platform APIs.
| name | code-review-graph-debug |
| description | Systematically debug issues using the code-review-graph knowledge graph. Use instead of manual file reading when tracing call chains, locating a bug's origin, or assessing impact radius of a suspected file. |
Use the knowledge graph to systematically trace and debug issues.
mcp_code-review-g_get_minimal_context_tool first with task="<your task>" to orient cheaply.mcp_code-review-g_semantic_search_nodes_tool to find code related to the issue.mcp_code-review-g_query_graph_tool with callers_of and callees_of to trace call chains.mcp_code-review-g_get_flow_tool to see full execution paths through suspected areas.mcp_code-review-g_detect_changes_tool to check if recent changes caused the issue.mcp_code-review-g_get_impact_radius_tool on suspected files to see what else is affected.#tool:read/readFile when the graph result is insufficient — prefer the graph for navigation and reserve file reads for inspecting specific implementation details.mcp_code-review-g_get_minimal_context_tool before any other graph tool.detail_level="minimal" on all calls. Only escalate to "standard" when minimal is insufficient.