Toolbox/Agent capabilities

RAG & retrieval

retrieval and knowledge layers for agents

10 repositories, most starred first

/

State of the category

infiniflow/ragflow infiniflow/ragflow is the clear category leader; before choosing, decide whether you want a full, production-ready agentic RAG engine (pick ragflow) or a narrower building block—crawler, indexer, vector store, or domain skill pack—you'll integrate yourself.

Flexible ingestion

Teams need robust, configurable crawlers and ingestors to turn messy web pages, books, and enterprise metadata into high-quality LLM-ready context, and many projects specialize just at that step.

→ crawl4ai, book-to-skill, llama_index, ktx

Traceable retrieval

For regulated, research, or high-stakes use cases teams prioritize retrieval methods that produce explainable, human-like organization and provenance rather than opaque nearest-neighbor hits.

→ PageIndex, llama_index

Lightweight search primitives

Some teams only need a minimal, embeddable search layer or local-first search to prototype or run offline without adopting a full RAG stack.

→ chroma, zvec-grep, PageIndex

Domain-specific skills

Organizations building vertical applications prefer curated, domain-aware skillpacks and mappings (e.g., cybersecurity workflows or book-to-skill conversions) so agents answer with practitioner-grade knowledge.

→ Anthropic-Cybersecurity-Skills, book-to-skill

Production safety

Production teams demand auditability, dry-run/confirmation workflows, and governance-friendly CLIs or reconciled semantic layers to safely automate retrieval-driven actions.

→ SearchCLI, ktx

Find your fit

Which one matches your setup?

Answer any of the questions — the shortlist updates as you go. Recommendations come from the capability passports below, nothing else.

Where should reviews happen?

Can code leave your infrastructure?

What matters most?

Model access?

Pick at least one answer to get a shortlist.

Side by side

Comparison matrix

Axes are extracted from each project's docs by our review pipeline; the maturity score is computed from stars, growth and commit activity — not an opinion. Click a column to sort.

	Runs in	Models	Context					Cost to run
infiniflow/ragflow ↗Leader ⭐ 92.5k +388/7d	Web app	BYOKOpenAIGemini	Whole-repo analysis	Re-ranking & citations	●●●●●	Self-hostable	●●●●●	Your API key (paid LLMs); self-host or use cloud
run-llama/llama_index ↗ ⭐ 52.6k +104/7d	CLIWeb appCI	BYOKOpenAIOllama / local	Whole-data ingestion	Rerankers + filters	●●●●●	Local model support	●●●●●	Your API key (or optional LlamaParse cloud)
VectifyAI/PageIndex ↗ ⭐ 39.3k +230/7d	CLIWeb app	BYOKOpenAI	Whole-repo analysis	Basic filters/config	●●●●●	Self-hostable	●●●●●	Uses your LLM API key (e.g. OpenAI); cloud service may be paid.
virgiliojr94/book-to-skill ↗ ⭐ 35.4k +697/7d	CLICoding-agent plugin	BYOK	Whole-skill access	Validation & rules	●●●●●	Local processing	●●●●●	Free, local processing; external model API costs apply
chroma-core/chroma ↗ ⭐ 29.5k +45/7d	CLIWeb app	BYOK	No repo context	Basic filters	●●●●●	Runs locally	●●●●●	Free to self-host; optional Chroma Cloud hosted service
zvec-ai/zvec-grep ↗ ⭐ 4.1k +46/7d	CLICoding-agent plugin	BYOK	Whole-repo index	Ranked + filters	●●●●●	Fully local	●●●●●	Free local use; external models may require your API key.
unclecode/crawl4ai ↗ ⭐ 85.9k +469/7d	CLIWeb appCoding-agent plugin	BYOKOpenAIOllama / local	Whole-site crawl	Filters (BM25/LLM)	●●●●●	Fully local possible	●●●●●	Free to self-host; Crawl4AI Cloud is pay-as-you-go (first $10 free promotion)
mukul975/Anthropic-Cybersecurity-Skills ↗ ⭐ 34.7k +400/7d	CLIIDECoding-agent pluginWeb app	AnthropicOpenAIGemini	Indexed frontmatter + selected skills	Matching + verification	●●●●●	Self-hostable	●●●●●	Free to clone; your model/API costs
Kaelio/ktx ↗ ⭐ 1.6k +3/7d	CLICoding-agent plugin	BYOKAnthropicOpenAIFixed provider	Whole-repo analysis	Dedup & human review	●●●●●	Self-hostable	●●●●●	Your LLM API key; ktx adds no extra usage billing
volcengine/SearchCLI ↗ ⭐ 1.2k	CLICoding-agent plugin	BYOK	Local-only	Dry-runs & confirmations	●●●●●	Cloud APIs with key	●●●●●	Volcengine account + optional LLM API key

At a glance

Capability profiles

Six axes, 0–5 each. The shape tells you the strategy: a wide hexagon is a generalist, a spike is a specialist. Showing the 8 most established — the rest are in the full catalog.

infiniflow/ragflow⭐ 92.5k

Combines a production-grade RAG engine with agent orchestration and template-based document understanding to provide an agentic context layer for LLMs.

run-llama/llama_index⭐ 52.6k

A data-first RAG framework focused on document agents and agentic OCR with extensive indexing, retrieval, and 300+ integrations.

VectifyAI/PageIndex⭐ 39.3k

Vectorless, reasoning-based hierarchical retrieval that builds a human-like table-of-contents tree for traceable, context-aware RAG without vector DBs or chunking.

virgiliojr94/book-to-skill⭐ 35.4k

Converts books and document collections into cross-agent, on-demand 'skills' with per-chapter files so agents answer from real content without dumping entire books into context.

chroma-core/chroma⭐ 29.5k

Provides a minimal, 4-function client API for fast in-memory prototyping plus an optional hosted Chroma Cloud for production-ready vector search.

zvec-ai/zvec-grep⭐ 4.1k

Unifies ripgrep, BM25, and vector search into a local-first search layer that serves both humans and AI agents.

unclecode/crawl4ai⭐ 85.9k

Open-source crawler that turns any website into LLM-ready Markdown with full browser control and optional hosted cloud + agent MCP integration.

mukul975/Anthropic-Cybersecurity-Skills⭐ 34.7k

Maps 817 structured cybersecurity skills to six major industry frameworks (e.g., MITRE ATT&CK and MITRE F3), providing practitioner-grade, cross-framework workflows for AI agents.