daidocs

daidocs is a plain-text code graph tool that measures code structure and token cost against Graphify.

Daidocs uses tree-sitter to parse code and write its structure down as plain text in a .Cai format, creating a readable code graph of functions and calls. Coding agents can query this index to find imports, callers, and dependencies with significantly fewer tokens than reading raw source files or using graph databases. The plain-text structure can be opened, grepped, and diffed in git, making it suitable for developers and coding agents needing efficient code retrieval. It compares favorably against Graphify on real repositories while offering transparent benchmarking data. Users can run the Apache-2.0 v4.4n engine locally or use the hosted dashboard.

daidocs at a glance

Best for: Software engineer
Runs on: Web
Pricing: Free forever on open Apache-2.0 engine; free with own API key or subscription; prepaid credit top-up for conversions; Gold supporter $25 credit
Released: October 2026
Category: AI coding tool
Website: daidocs.com

Key features

Parses code structure into plain-text graphs
Uses tree-sitter parser for code analysis
Stores code graphs in flat .Cai files
Provides hybrid lexical and MiniLM retrieval
Supports local execution via Apache-2.0 engine
Offers cloud conversion and dashboard options

Use it for

Querying code imports and dependencies
Finding function callers and callees
Retrieving code definitions and symbols
Running token cost and accuracy benchmarks

Companions

Tools that articles about daidocs recommend using with it, each with its job.

Graphify — Built to work with Graphify.
Claude Code — Installs as a skill so the agent queries the graph before grepping.
Codex — The installer wires the same graph into Codex.
Cursor — Auto-detected and configured alongside Claude Code.
OpenCode — Supported for mixed-agent setups.
Obsidian — Open the output as a vault to see how the code clusters.

Alternatives to daidocs

jevgrep — Reduces coding agent costs by providing targeted, relevant repository context.
CodeGraph — It eliminates file crawling, reducing AI agent tool calls by 88% and cutting costs.
ccusage — Tracks token usage and costs locally across multiple coding agent CLIs.
jCodeMunch MCP — Cuts AI code-reading token costs by over 95 percent via tree-sitter symbol indexing.
CodeSight — Reduces AI token usage significantly and maps projects instantly.
TokenTracker — Monitor AI spending across 43 coding tools in one local, private dashboard.
Axiomcodegraph — Resolves calls from types rather than syntax for higher accuracy.
Code Review Graph — Provides a median 63x context reduction while keeping all code processing local.
Tencent CodeBuddy — Offers multi-model support and context-aware coding assistance across many IDEs.
jscpd — A native Rust binary that runs fast without a Node.js runtime and integrates with AI agents.
Cost X-Ray — It prices usage below the tool level by analyzing raw API request bytes.
DeepSource — Automated code reviews that combine static analysis with AI agents.

All daidocs alternatives