Cowrie Research

seda-primary-sources-2026-04-25

4 min read

SEDA primary-source snapshot

Fetched: 2026-04-25 UTC

1) Official homepage / positioning

Source: https://seda.xyz/

Selected excerpts from the official site:

Access programmable oracle infrastructure in minutes. Powering your favorite markets.

Access bespoke oracle infrastructure to fit your needs.

Access 24/7 pricing for any trad-Fi asset

Additional signals from the homepage:

  • The site positions SEDA as infrastructure for custom data access and not only as a commodity price-feed provider
  • Official copy emphasizes bespoke oracle methodologies, support for any API/data source, and tradfi-style asset coverage rather than a narrow crypto-only feed set
  • The homepage also points directly to docs, token docs, and GitHub, which makes the technical source of truth fairly discoverable

2) Introduction / developer-facing docs

Source: https://docs.seda.xyz/home Source: https://docs.seda.xyz/home/introduction-to-seda.md?ask=What%20does%20SEDA%20actually%20provide%20to%20developers%2C%20how%20does%20its%20oracle%20model%20work%2C%20and%20what%20product%20surfaces%20or%20delivery%20options%20are%20documented%3F

Selected excerpts from the docs:

SEDA gives developers programmable oracle infrastructure through Oracle Programs.

Those programs let you define what data to fetch, which sources to use, and how to filter, aggregate, and return it.

SEDA Core — permissionless onchain oracle infrastructure.

SEDA Fast — an ultra-low-latency endpoint for Oracle Programs.

Additional signals from the docs:

  • The docs make the product split unusually legible: SEDA Core for onchain delivery and SEDA Fast for lower-latency API/WebSocket access
  • This looks more like a programmable execution layer for oracle logic than a static feed catalog
  • The “what data / which sources / how to aggregate / how to deliver” framing is a strong clue for how the project should be categorized

3) Oracle Programs / architecture docs

Source: https://docs.seda.xyz/home/technology/seda-network-architecture/oracle-programs

Selected excerpts from the docs:

SEDA introduces the concept of Oracle Programs, containers that consist of protocol-defined instructions that dictate how their data requests should be executed.

Oracle Programs allow the builder to customize a feed for any data type.

Once an Oracle Program is deployed to the SEDA Network, it is stored as a WASM Binary and assigned an Oracle Program ID.

Additional signals from this page:

  • The architecture docs are the clearest primary-source explanation of SEDA’s core primitive
  • The public/private-source distinction is important because it shows SEDA is designed for more than open-web price polling
  • The WASM-binary / Oracle Program ID model suggests a durable programmable artifact, not just an API endpoint with marketing around it

4) SEDA Fast / agent-oriented docs

Sources:

Selected excerpts from the docs:

SEDA FAST is a specialized oracle framework built on top of SEDA Core tooling designed for low-latency access to Oracle Programs across any network.

SEDA is programmable oracle infrastructure for autonomous agents.

It provides a way for agents to request data, apply deterministic policy, and deliver results onchain through Oracle Programs.

Additional signals from these pages:

  • SEDA treats agent-facing workflows as first-class documentation, not as an afterthought blog post
  • The docs explicitly frame deterministic policy execution, machine-to-machine requests, and copyable runbooks as part of the product surface
  • This is a useful long-tail discovery pattern: oracle/data projects with AI-agent runbooks often reveal more real operating surface than their homepage alone

5) Audits and public code surface

Sources:

Selected excerpts/signals from official materials:

Direct Report Link

Direct PDF Link

Starter kit for building Oracle Programs with the SEDA-SDK.

SEDA chain contracts written in CosmWasm

EVM contracts to interact with the SEDA network

Additional signals from the audit and repo surface:

  • The docs explicitly link to public audit materials, including a Sherlock report and PDF, which supports a production-readiness reading of the project
  • The GitHub org exposes the chain, starter kit, contracts, verification components, and example programs rather than only marketing collateral
  • The llms.txt index is especially useful because it surfaces agent modules, data-proxy docs, node-operation docs, prover-contract pages, and audit pages in one machine-readable corpus

Notes

  • This pass did not locate a classic standalone SEDA whitepaper or litepaper as a primary canonical PDF, even though the architecture docs are substantial and reference a broader technical framing.
  • The strongest primary materials were the homepage, introduction and architecture docs, SEDA Fast docs, autonomous-agent docs, audit pages, llms.txt, and public GitHub organization.
  • SEDA appears best categorized as programmable oracle and custom data-execution infrastructure with both onchain and low-latency offchain delivery surfaces.

Graph View