# parcadei/llm-tldr

95% token savings. 155x faster queries. 16 languages.  LLMs can't read your entire codebase. TLDR extracts structure, traces dependencies, and gives them exactly what they need.

Repository: https://github.com/parcadei/llm-tldr
Canonical: https://ross.abutalabs.com/products/llm-tldr
Language: Python
License: AGPL-3.0
License Family: copyleft
Archived: true
Last push: 2026-01-17T09:59:21+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 62, release rhythm 35, longevity 16
- inputs: {"age_days": 236, "days_push": 228, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1170, forks 114 (observed 2026-08-28T04:03:51.083582+00:00)

## What it is
A Python CLI tool and daemon that indexes codebases into a 5-layer analysis stack (AST, call graph, control flow, data flow, program dependence) using tree-sitter, plus a FAISS semantic index for natural-language code search. It serves token-efficient, LLM-ready code context to AI agents, claiming 95% token savings and millisecond queries across 16-17 languages.

## Use cases
- give my coding agent only the relevant parts of a large codebase
- reduce token usage when feeding code to Claude or GPT
- find where JWT validation is implemented via semantic code search
- trace what affects a specific line before refactoring
- generate LLM-ready summaries of project structure and dependencies
- index a multi-language monorepo for AI-assisted development

## When to choose
- you use AI coding agents on codebases too large for the context window
- you want fast semantic and structural code queries with a persistent daemon
- you need call graph, data flow, or program slicing analysis across many languages

## When to avoid
- you need a full IDE-grade static analysis or type checker
- your project is small enough to fit directly in the LLM context
- you require a permissive license (AGPL-3.0)
- you need deep analysis for languages outside its supported set

## Facets
- artifact type: cli-tool
- maturity: active
- function: parser, search-engine, rag, mcp, developer-tools
- domain: developer-tools, large-language-models, parsers, apis
- platform: cli, python, cross-platform
- tags: static-analysis, tree-sitter, call-graph, token-optimization, code-indexing, semantic-search, context-window, ai-agents

## Member repositories
- parcadei/llm-tldr (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:51.083582+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:28:38.507367+00:00, confidence not recorded.
  - readme: https://github.com/parcadei/llm-tldr (fetched 2026-08-28T04:03:51.083582+00:00, sha a8be3df32e5d)
  - registry_pypi: https://pypi.org/pypi/llm-tldr/json (fetched 2026-08-29T12:34:29.550351+00:00, sha fe6a064e1a65)
- Data as of 2026-08-30T08:39:29.467469+00:00.