Fodda's Catalog is Now Agent-Readable (Open Knowledge Format)
News / Integration — Aug 2, 2026
By Piers Fawkes · Founder, Fodda & PSFK · 4 min read
Agents cannot buy what they cannot find.
If an AI agent needs consumer trends, category intelligence, or market data, how does it know where to look? Until now, discovery was an ad-hoc mess of hardcoded prompts and custom tool declarations.
In June 2026, Google published the Open Knowledge Format (OKF) as an open, vendor-neutral standard. OKF packages domain knowledge into cross-linked markdown concept files with simple frontmatter metadata. An agent reads these concept files to understand what a knowledge source covers, how queries are structured, and when to invoke it.
Fodda already produced these exact knowledge structures internally. Today, we emit our entire catalog of 250+ knowledge graphs, expert collections, and institutional sources as a live OKF bundle at GET /v1/catalog.okf.
The Menu vs The Meal: The Freshness Moat
A downloaded file freezes the moment it lands on a disk. Consumer intelligence, economic indicators, and retail trends age the second you stop pulling from the live source.
That is why our OKF bundle is strictly metadata-only. The OKF bundle is the menu. The live Fodda API is the meal.
The public bundle carries routing hints, entity types, coverage bounds, topic counts, offering lists, and payment terms. It never contains the trend or evidence content itself. By keeping the menu public and the meal live, nothing on the menu goes stale. Your agent reads the menu to discover what exists, then queries the live API to retrieve real-time, cited evidence.
Payment Travels in the File
How does an agent pay for data without a human logging into a dashboard? Every concept doc in the Fodda OKF bundle tells the agent the exact price and payment mechanism before a query is sent:
---
id: retail
name: PSFK Retail Intelligence
curator: PSFK
graph_type: domain
price_per_call: 0.50
typical_calls: 15
payment_methods:
- mpp_spt
- api_key
- enterprise_oidc
mpp_402_endpoint: https://api.fodda.ai/v1/search/domain
---When an autonomous agent discovers a concept doc, it sees the cost ($0.50 per API call via Shared Payment Tokens) and the endpoint. If it lacks an API key, it hits the endpoint, receives an HTTP 402 payment prompt, pays via SPT/Stripe, and completes the research task in a single automated pass. Account and subscription users draw on their plan credits automatically.
Skills as Well as Sources
The OKF bundle surfaces Fodda's Use Case Skills alongside data sources. An agent discovers not just what Fodda knows (such as retail signals or FRED economic indicators), but what workflows it can commission, including:
- Brand Intelligence Profiling: Mapping a brand's footprint against expert trend signals.
- Deep Research Streams: Multi-step synthesis across expert collections and earnings calls.
- Earnings Intelligence: Detecting analyst pressure and executive deflection across 500+ public companies.
The Citability Side Effect
There is a structural benefit to emitting structured, agent-readable knowledge: machine citability. When AI search crawlers index structured OKF concept files marked up with schema.org (Dataset, DefinedTerm, and Organization), they cite Fodda's expert sources in LLM responses rather than returning flat web summaries. The menu is also how the machines learn that Fodda exists.