Exa — Semantic Search

Trust Score 64/100

Perform semantic (meaning-based) search over documents or collections to find similar content and neural search results.

triggers:semantic searchfind similar documentsneural searchretrieve by meaningRAG retrieval

GitHub SKILL.md

What it does

Exa adds semantic search capability to an agent: it performs meaning-based retrieval (not keyword matching) to return documents, passages or results similar to a query. Typical outputs are ranked text results useful for research, RAG pipelines, and content discovery.

When to use it

Use Exa when you need to find related documents by concept rather than literal terms, surface similar examples, or augment a retrieval-augmented generation workflow with semantically relevant sources. Good for ad-hoc research, summarization, and building search UIs.

What's included

Scripts: none embedded in the skill (has_scripts=false) — integration shown via CLI workflow examples.
References: none listed.
Instructions: the SKILL.md includes example commands to run the skill via Refly (skill run -> get RUN_ID -> toolcalls) and shows how to extract text content from workflow toolcalls. The input schema expects a JSON object with a search_query and num_results.

Compatible agents

Inferred compatibility: Claude Code / Refly workflows and generic agent orchestration platforms that can call external skill workflows. The skill is authored for Refly's ecosystem and expects the Refly CLI/workflow runtime.

Audit Summary

The Exa skill provides semantic search via Exa AI, wrapped in the Refly platform CLI. No bundled scripts — it's a single SKILL.md that calls refly CLI commands. Instructions are thin: just a bash command pattern with no error handling or edge cases. The skill depends entirely on the proprietary refly CLI and a Refly account, making it unusable outside that ecosystem. Input field naming is inconsistent (search_query vs query).

Watch Out

Requires refly CLI installed and authenticated
Depends on Refly platform account and workflow
Input JSON example uses 'search_query' but execution snippet uses 'query'
References external base skill at ~/.claude/skills/refly/SKILL.md which may not exist

Notes

Platform-specific skill with no standalone value. SKILL.md is a thin wrapper around refly CLI commands. No security concerns — no scripts, no credentials, no destructive commands. Main issues are low code quality (minimal instructions, no error handling) and limited usefulness outside the Refly ecosystem.

Information

Repository: refly-skills
Stars: 186
Installs: 0

Trust Score

Overall64

Security92

Code Quality35

Architecture40

Usefulness45

Related Skills

Retriever (Membrane CLI)

Integrate Retriever search and discovery via the Membrane CLI — manage connections and run queries across Retriever data sources.

arxiv2md

Convert arXiv papers into clean, LLM-ready Markdown via a REST API or CLI for fast reading, summarization, and downstream processing.

ClawPod / Massive Unblocker

Bypass anti-bot restrictions and fetch rendered HTML or structured search results via Massive's Unblocker API (handles CAPTCHAs, JS rendering, geo-restrictions)

Retriever (Membrane) Integration

Integrate with Retriever via the Membrane CLI to search, query, and proxy Retriever data (documents, snippets, queries) with built-in auth and pagination.

Generate Multilingual Sentences with Contextual Definitions

Produce multiple example sentences using a target word in varied meanings, translate them into a target language, and show the contextual meaning of the word fo

Humanize AI Text

Rewrite AI-generated text into more natural, human-like prose using the HumanizerAI API, with before/after scores and credit tracking.

Algolia Cost Tuning

Practical guide to reduce Algolia billing by auditing records/requests, switching to virtual replicas, multi-query usage, caching, and cleanup strategies.

Memory Search

Search past coding sessions and observations using natural-language queries to retrieve decisions, notes, and context.

Back to Skills