Perplexity
Perplexity AIAn answer engine that cites what it used.
Scores retrieval quality on a fixed question set.
Answering whether a retrieval change made the system better
Building the pipeline — it evaluates, it does not index
Runs a fixed set of questions against a retrieval pipeline and reports two numbers: whether the expected document appeared, and whether the answer cited it. Both matter, because a system that retrieves the right document and cites a different one is worse than it looks in a demo.
Clone the repository, then copy the skill folder into your agent's skills directory — ~/.claude/skills/ for Claude Code. Skills are plain files: no build step, no daemon.
npx skills add rag-evalRead the SKILL.md before you install. A good skill states when it applies, what it must not do, and how to verify the result. A skill that only lists adjectives will change the agent's tone without improving its output — skip those.
An answer engine that cites what it used.
Search API shaped for agents: results plus a short synthesised answer.
Enterprise search and an assistant that answers from your company's own documents.
Reads channels and groups through the Bot API: recent messages, search, and posting. The distinction that matters is bot token versus user session — a bot sees what …
Wraps the YouTube Data API for search, channel and video metadata, and captions where available. The transcript path is the useful one for research; anything that re…
Searches subreddits and reads threads with comments through the Reddit API. It is a reasonable research input and a poor automation target: the platform's rules and …