What problem does it solve? Debugging iOS memory leaks requires capturing memory graphs from running simulator processes and interpreting Apple's leaks output, which is tedious and error-prone when done manually. This Skill automates memgraph capture, leak summarization, and ownership-path analysis so you can prove and fix retain cycles with before/after evidence. ## Core Features & Use Cases - Memgraph Capture: Capture a .memgraph from a running iOS simulator app by UDID and bundle identifier using the capture_sim_memgraph.sh script. - Leak Summarization: Parse leaks output into a Markdown report with top leaked types, images, traceTree excerpts, and grouped leak trees via summarize_memgraph_leaks.py. - Root-Cause Workflow: Follow structured rules for identifying app-owned leaked types, proving retain cycles with ownership paths, and verifying fixes with before/after captures. - Use Case: After driving a flow that should release objects, capture a memgraph, summarize the leaks, trace the retaining edge for an app-owned type, apply the smallest fix, and recapture to prove the leak disappeared. ## Quick Start Use the ios-memgraph-leaks skill to capture a memgraph from my running simulator app and summarize which objects are leaking.