Phony

Availableapp.erp.io/phony/knowledge

Knowledge

The agent knows what you have given it. Almost always that is your website, and almost always the interesting question is which parts of it.

Source
Your site, plus documents
Discovery
Sitemap first
Page cap
Applies silently
Scope
Per agent

How a site is ingested

The crawler asks for your sitemap

It requests /sitemap.xml first and only follows links if there is none. A site with a good sitemap is ingested completely; a site without one is ingested as far as its internal links reach.

Pages are fetched and cleaned

Navigation, footers and boilerplate are stripped so the agent is not repeating your menu at people.

Content is indexed

And scoped to the agent you are building.

A brief is compiled

A short account of what the company is. This is a separate step. See Agents.

There is a page cap, and hitting it looks like success

A crawl stops after a fixed number of pages. A site larger than that is ingested partially, and the run still reports as complete — so a large site can end up with an agent that confidently knows the first section of the website and nothing about the rest. If your site is big, check the page count against what you expect rather than trusting the status.

A sitemap index is not a sitemap

Large sites often publish a sitemap index that points at several child sitemaps. If the crawler is pointed at the index and does not descend into the children, it finds almost nothing — which presents as a nearly empty corpus on a site with thousands of pages. Point it at a specific sitemap if the index yields nothing.

app.erp.io/phony/knowledge
Knowledge

Reception agent · 46 pages · recompiled 2 days ago

Recompile briefAdd source
SourceTypePagesLast ingestedState
northwind.exampleWebsite462 days agoComplete
northwind.example/helpWebsite182 days agoComplete
Price list 2026.pdfDocument49 MarComplete
docs.northwind.exampleWebsite2002 days agoCapped
The knowledge screen. Sources, page counts and when each was last ingested.

Scoping

Knowledge is scoped to an agent. If you run agents for several brands — or, as an agency, for several clients — each agent must be scoped to its own sources. Getting this wrong produces an agent that answers a question about one client using another client's content, which is the worst possible failure in this module and is not always obvious from a transcript.

  • Scope every source to the agent that should have it, at the moment you add it.
  • After adding sources for a second brand, ask the first agent a question only the second brand could answer. It should not know.
  • Agency workspaces make this explicit — see Agency.

Diagnosing a bad answer

SymptomLikely causeCheck
Vague about what the company doesEmpty or stale brief.Recompile the brief.
Knows the homepage, nothing deeperNo sitemap, shallow link crawl.Publish a sitemap, re-ingest.
Knows a fraction of a large sitePage cap.Compare page count to reality.
Almost nothing ingestedSitemap index not descended.Point at a child sitemap.
Answers with another brand's factsScoping.Check which sources the agent has.
Confidently wrong about a priceStale content, or a PDF that was superseded.Re-ingest and remove the old document.

What this does not do

No live lookups

The agent answers from ingested content. It does not fetch a page during a conversation.

No authenticated crawling

Content behind a login is not ingested.

No automatic re-crawl

Re-ingestion is triggered, not scheduled. Content changes do not propagate on their own.

No per-page exclusion rules

Scoping is per source, not per URL pattern.

Questions

Can we upload documents?

Yes — price lists and FAQs are the common ones.

How often should we re-ingest?

Whenever the site changes materially. Nothing does it for you.

Does it use content from our other modules?

No. Knowledge is what you give this module.