Back to Homepage
Palladium Labs open-sources Chainlink oracle kit for Canton developers
TECHNOLOGYECOSYSTEM

Palladium Labs open-sources Chainlink oracle kit for Canton developers

canton-oracle-kit brings together signed-price verification, freshness checks and an example updater. Palladium says the underlying oracle path is used by Alpend on MainNet, with operator permissions central to the design.

4 min read

Palladium Labs has open-sourced canton-oracle-kit, a set of Daml contracts and supporting tools that helps Canton applications verify Chainlink price reports and keep usable prices on the ledger.

The team announced the release on November 18, describing it as the oracle path used by Alpend on Canton MainNet. Version 1.0.0 includes a report-verification contract, a price store, an example updater and documentation for deployment and operations. Palladium’s code is available under the Apache 2.0 licence. Release announcement, public repository

For builders, the intended benefit is less repeated integration work. A lending market needs prices to value collateral and calculate borrowing capacity. Getting those prices into an application involves checking their authenticity, age and suitability for the transaction using them.

How a signed report becomes a usable price

Chainlink Data Streams delivers market data in reports signed by its oracle nodes. Chainlink’s Canton integration provides contracts that check those signatures against a Chainlink-issued configuration containing the authorised keys. Chainlink’s integration guide

Palladium’s kit connects that verification step to a price store. An off-ledger program, called the pusher, fetches a signed report and submits it to a Daml feeder contract. The feeder calls Chainlink’s verifier, decodes the report, checks its validity window and passes the price to the store.

Applications can then read the stored price within their own contract logic. The repository includes an example that values a holding by multiplying its quantity by the current price, showing how the data can feed into a financial calculation. Architecture and example

A pusher fetches a signed Chainlink Data Streams V3 report. On Canton, the feeder calls Chainlink’s verifier and checks report validity. The price store checks freshness, ordering and large changes. Application reads require operator authority.The price-ingestion path in canton-oracle-kit. The trust model gives the application operator administrative and direct-write authority.

The kit currently supports V3 crypto reports. Developers also need Data Streams access and a Chainlink-issued verifier configuration for their Canton party. The example pusher targets Node.js 22, while the published compatibility table lists Daml SDK 3.4.11 and Canton 3.4.x. Compatibility and onboarding

Rules for stale prices and sharp moves

The price store applies several checks before accepting an update. Prices must be positive, observations must fall within the configured age limit, and the observation timestamp must be newer than the active price’s timestamp.

It also puts a price move above a configured threshold into a pending state. A subsequent submission can activate the new value when it corroborates the pending price within the required freshness window. During that wait, the previous active price remains available subject to its age limit.

That introduces a delay during genuine sharp market moves. Applications therefore need to choose their freshness and deviation settings around the decisions those prices support, particularly collateral valuation and liquidation. Price-store rules

The pusher includes controls for updates triggered by price movement or elapsed time, alongside monitoring for failed fetches and prices that stop refreshing. These give developers a starting point for operating the feed after deployment. Example pusher

The operator still has a central role

The design gives the pusher and the application operator different permissions. The pusher submits reports through the verification path. The operator controls the trusted verifier configuration, feed mappings and administrative changes.

The operator also retains authority to write prices directly or recreate the store. The kit’s trust model therefore places responsibility on operator key controls and monitoring the transaction history to confirm that price changes followed the intended verification path. Documented trust model

Price reads are operator-controlled too. A consuming application needs the store operator’s authority in its Daml contract flow. Developers must also track the current store identifier, which changes with each update. Integration limitations

Production experience and the audit scope

Palladium says Alpend uses the oracle path in production. That follows the Chainlink Data Streams integration we covered in May, when Alpend and Raven Market described pricing uses in credit and options markets. Our earlier Chainlink coverage

The repository links to Halborn’s Alpend assessment. That report covers an Alpend code snapshot reviewed in June, including its Chainlink oracle integration, and records all reported findings as addressed. The current kit separately documents live-participant testing of initial verifier validation and verifier rotation as outstanding work. Halborn’s assessment, kit limitations

The release also continues Palladium’s support for Canton developers. In June, we covered its Genesis Fund for applications, infrastructure, tooling and research. The oracle kit makes one specific part of that engineering work available for other teams to inspect, adapt and integrate into their own applications. Our Genesis Fund coverage

Have a question about this story?
Ask CantonNews AI for context, its impact on the Canton ecosystem, or the latest related news.
Explore Further
Developer ResourcesExplore Daml, CIP-56 & Zenith →