MCP gives coding agents tools. Deslop gives them live repository memory: call find-similar before writing, then reuse the code that already exists.
Read Article arrow_forwardThe Manuscript · Field Notes
Blog
Notes on duplicate detection, AI-era codebases, and the decisions behind Deslop.
Coding agents generate Flutter widgets, repositories, and state-management classes at a pace no reviewer can keep up with. Flutter's verbosity hides the copies. Here is what to check, why line-based tools miss it, and how to deduplicate Dart structurally.
Read Article arrow_forwardAI-generated code can multiply duplicate logic before review catches it. This guide explains what to check, why duplicate code becomes technical debt, and where to read the full Deslop research background.
Read Article arrow_forwardCoding agents generate plausible code faster than humans can review it. They also duplicate it faster. Here's why Deslop treats that as the defining problem of the era, not a side-effect.
Read Article arrow_forwardDeslop ranks clusters by clone_node_count × (cluster_size − 1) × log2(1 + spanned_bytes). Every decision in the tool flows from that one line. Here's why it's not configurable.
Read Article arrow_forwardDeslop parses every language with tree-sitter. No regex, no line-matching, no heuristics. Here's why that constraint matters more than any feature the tool ships with.
Read Article arrow_forward