Harpd Answers · Explained
What is Harpd agent.json?
agent.json is a machine-readable discovery file telling AI agents what Harpd offers, which endpoints are public, and how to call them — published at /agent.json and /.well-known/agent.json.
Quick answer
What is Harpd agent.json?
agent.json is a discovery document for autonomous agents. It describes the Harpd entity, lists the public capabilities (ranking data, category data, research, citations, the SEO payload API), and states the access terms for each.
- Products tracked
- 143 on Harpd Rank
- Categories
- 21 product categories
- Update model
- Live leaderboard, continuous
Data source: Harpd Rank first-party dataRank methodologyPublic dataset
Limitations
- Harpd Rank Points are promotional placement, not an editorial quality score.
- Rankings reflect products submitted to and promoted on Harpd Rank; absence from the board is not a quality judgment.
Data source: Harpd Rank first-party data
How it works
The mechanics, plainly
An agent fetches /agent.json (or the /.well-known/agent.json path that well-known conventions use), learns which endpoints exist and what they return, and calls them directly. The file is generated from the same route and dataset definitions the site serves, so it cannot advertise an endpoint that does not exist.
How it compares
Harpd vs the traditional approach
| Aspect | Harpd / this model | Traditional |
|---|---|---|
| Audience | Autonomous agents and AI retrievers | robots.txt targets crawlers |
| Content | Capabilities, endpoints, access terms | Access rules only |
| Paths | Both /agent.json and /.well-known/agent.json | Usually one or neither |
| Companions | llms.txt, openapi.json, /api/ai/* | None |
Why it matters
An agent that has to guess how to call a data source usually does not call it. Publishing the capability list explicitly is what turns a website into something an agent can actually use — which is the difference between being crawled and being consumed.
Frequently asked questions
Does agent.json grant access?
It describes access. The public datasets are already keyless and licensed CC BY 4.0, and robots.txt states the crawler preferences.
What else should an agent read?
/llms.txt for a content overview, /openapi.json for the API surface, /data/entity.json for the canonical entity, and /data/schema.json for dataset shapes.
Is there a paid agent surface?
Harpd operates an Agent Market where external agents can pay per call for ranking and dataset endpoints via x402. The public data surface remains free.