llamaindex

Data framework for building LLM applications with RAG. Specializes in document ingestion (300+ connectors), indexing, and querying. Features vector indices, query engines, agents, and multi-modal support. Use for document Q&A, chatbots, knowledge retrieval, or building RAG pipelines. Best for data-centric LLM applications.

Quality

78%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./14-agents/llamaindex/SKILL.md

Quality

Content

65%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A highly actionable reference packed with executable, copy-paste-ready examples, but it is verbose and monolithic: marketing metrics, version numbers, comparison tables, and many inline topics inflate the body when they belong in separate reference files. Workflow steps lack explicit validation checkpoints.

Suggestions

Remove or relocate the GitHub-star/contributor metrics, performance-benchmarks table, and LlamaIndex-vs-LangChain comparison out of the main body; these are marketing/conceptual padding rather than actionable guidance.

Move version numbers (e.g., 'v0.14.7') and other time-sensitive data into a clearly labeled 'Version notes' or 'Deprecated' section per the conciseness guideline, or drop them entirely.

Split the long inline sections (vector-store integrations, multi-modal RAG, evaluation, advanced RAG patterns) into additional reference files alongside the existing three, keeping SKILL.md as a lean overview with well-signaled links.

Dimension	Reasoning	Score
Conciseness	The body is mostly executable code, but it is padded with marketing metrics ('45,100+ GitHub stars', '1,715+ contributors', 'v0.14.7'), a LlamaIndex-vs-LangChain comparison table, a performance-benchmarks table, and repetitive vector-store examples, plus time-sensitive version numbers not placed in a deprecated section. Not 1 because the bulk is concrete code rather than conceptual explanation; not 3 because the metrics, comparison, and overall length could be cut.	2 / 3
Actionability	Provides numerous copy-paste-ready, fully executable Python examples spanning ingestion, indexing, querying, agents, vector stores, and evaluation, matching the top anchor for specific, complete examples.	3 / 3
Workflow Clarity	Sequences are present (Quick start load→index→query, and the 'Complete RAG pipeline' common pattern), but there are no explicit validation checkpoints or feedback loops, and the evaluation section is not integrated as a verify step in the pipeline. Not 1 because the sequences are clear; not 3 because checkpoints are missing or implicit.	2 / 3
Progressive Disclosure	The References section links three real, one-level-deep files (query_engines.md, agents.md, data_connectors.md) with clear signaling, but the ~560-line body itself is largely monolithic, with inline topics (vector-store integrations, multi-modal, evaluation, benchmarks, comparison) that would be better split into separate reference files. Not 1 because the references are real and well-organized; not 3 because content that should be separate is inline.	2 / 3
	Total	9 / 12 Passed

Description

92%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description that covers concrete capabilities and explicit use-when triggers with natural keywords. Its main weakness is overlap risk with general-purpose LLM framework skills, since the broad 'LLM applications / chatbots' framing competes with alternatives like LangChain.

Suggestions

Tighten the 'when' clause to RAG-specific triggers (e.g., 'Use when the task is retrieval-augmented generation or document Q&A over a private corpus') to reduce overlap with general LLM/agent skills.

Drop the broad 'building LLM applications' and 'Best for data-centric LLM applications' framing, which is too generic and increases conflict risk with LangChain-style skills.

Consider adding a 'Prefer LangChain when...' boundary to make the niche explicit rather than relying on the body's comparison table.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions and capabilities such as 'document ingestion (300+ connectors), indexing, and querying' plus 'vector indices, query engines, agents, and multi-modal support', matching the score-3 anchor for naming several specific concrete actions.	3 / 3
Completeness	Explicitly answers what ('Data framework for building LLM applications with RAG...') and when with an explicit trigger clause ('Use for document Q&A, chatbots, knowledge retrieval, or building RAG pipelines'), satisfying both halves required for the top anchor.	3 / 3
Trigger Term Quality	Includes natural terms a user would actually say such as 'RAG', 'document Q&A', 'chatbots', 'knowledge retrieval', and 'building RAG pipelines', giving good coverage of natural trigger keywords.	3 / 3
Distinctiveness Conflict Risk	The RAG/document-ingestion niche is identifiable, but broad framing like 'building LLM applications', 'chatbots', and 'knowledge retrieval' overlaps with general LLM-framework skills such as LangChain, which the description itself names as an alternative. Not 3 because of that overlap risk; not 1 because the RAG/data niche is still distinct.	2 / 3
	Total	11 / 12 Passed

Validation

87%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (570 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: Orchestra-Research/AI-Research-SKILLs
Path: 14-agents/llamaindex/SKILL.md
Commit: 773a529

Reviewed: about 13 hours ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.