// context, not tokens
YOUR LLM
CAN'T READ
ALL YOUR CODE
TLDR extracts structure, traces dependencies, and gives LLMs exactly the context they need. 95% fewer tokens. 17 languages.
5 Layers of Analysis
AST
Code structure — functions, classes, imports. What exists and where.
Call Graph
Who calls what. Cross-file dependency tracing.
Control Flow
How execution branches. Loops, conditionals, error paths.
Data Flow
How values propagate. Variable assignments, transformations, returns.
Program Dependence
What affects what. The minimal slice needed for any change.
Semantic Search
Find code by what it does, not what it's called. FAISS-backed embeddings.
Quick Start
$ pip install llm-tldr
$ tldr warm .
$ tldr context main --project .
# Function: main (src/app.py:42)
## Calls: init_db, start_server, register_routes
## Called by: __main__
## Data flow: config -> db_url -> engine
17 Languages
Python, TypeScript, Go, Rust, Java, C, C++, C#, Ruby, PHP, Kotlin, Scala, Swift, Lua, Elixir, Haskell, OCaml.