Decoder – GPS Navigation for Codebases

Name: Decoder – GPS Navigation for Codebases
Availability: InStock
Author: maryamtb

by maryamtb·Feb 13, 2026·1 point·1 comment

Visit Project View on HN

AI Analysis

●●●BangerBig BrainSolve My ProblemWizardry

Call graph traversal for LLMs cuts token usage vs. iterative grep and file reads.

Strengths

•AST+SQLite architecture surfaces dead code and conditional context automatically
•MCP server integration lets Claude traverse graphs directly instead of token-wasteful file reads
•VS Code extension + CLI + server means three UX layers for different personas

Weaknesses

•Python-only limits audience in polyglot teams
•No TypeScript/Go/Rust support when those langs have equally complex call chains

Post Description

I built decoder to visually trace and step through call chains across a codebase, without having to sift through several files, and with visibility into dead code you might otherwise never notice.

Decoder parses the Python AST to build a call graph stored in a local SQLite file, then lets you trace full call chains and see execution context (conditionals, loops, try/except). I built this first as a VS Code extension, but saw the value in giving LLMs that same visibility and added an MCP server. Instead of iterative grep and file reads, an LLM can traverse the call graph directly — which cuts down on token usage and back-and-forth significantly.