
By Prices API RFP - CTX
> Update Note - 27 February 2026: We've made some updates and clarifications in response to reviewer feedback. Changes address: (1) Horizon and RPC dependency risks - we will self-host all Stellar node infrastructure across our colocated and cloud environments with full redundancy, and will evaluate Galexie and the Composable Data Platform, our general approach is to always go low level and we'll either read from ledgers, use existing tooling (CDP) or even implement the protocol itself; (2) an itemized budget breakdown; and (3) concrete implementation plans for all required oracles and liquidity venues. We have no plans to rely on third-party hosted node providers - our existing infrastructure stack and colocated hardware give us the foundation to run everything ourselves. All eight required sources (Chainlink, Redstone, Band Protocol, Reflector, Soroswap, Aquarius, SDEX, Blend) are planned deliverables; integration approaches for each are now detailed in the technical architecture document.
Additionally I would like to make the following clarifications:
- We're committed to delivering the full RFP & Freighter RFQ to above and beyond there specification. There will be some discovery along this build process and we'll definitely evaluate the full composable data platform and Galaxie and document our findings and will use them where suitable (this may involve forks and/or pull requests too).
- We have no intention of relying on third party providers, this involves more risk and one of the core benefits of our experience is in our infrastructure knowledge and existing stack; we will be deploying multiple nodes ourselves (colocation and cloud) with full failover, everything will be designed from the ground-up to keep working as a reliable long term pricing partner with graceful degradation and rapid restoration in the event of any issues (network issues like the P23 deployment stability issues, AWS/cloudflare outages, Oracle freeze and so on). If we do integrate third party data providers (Data lake/nodes/otherwise) that will be more as a redundancy to our main self-hosted approach.
- Before recent events I highlighted my concerns around supply chain attack risk to a couple of team members, this is a critical piece of infrastructure that needs to be built right, whilst we will have it in production per the timeline a third party audit of our methodology around pricing would be very logical - especially if we then additionally publish data on chain as a secondary oracle to Reflector down the line. As we've built for cross-chain decentralized exchanges before we know these attack vectors well so I am confident we'll approach this right but having additional eyes on the project would be very advisable and we would gladly make use of the audit bank support. That doesn't mean this wont be immediately usable per these timelines but I would recommend any financial protocols relying on this data as a primary source (outside of pure 'visual' use cases) to await audits as a generalized precaution. This applies to any infrastructure like this and is general good sense.
- Budgets for development hours have been added to each tranche & milestone, broken down by hours weeks and rate across 3 full-time developers to achieve this to the highest standard in 10 weeks. Unspent development time will go towards additional infrastructure costs (taking us beyond the 6 months budget detailed below) or post-launch updates.
- Infrastructure & tooling (including API costs for coinmarketcap/coingecko/forex providers/etc) is budgeted at $30,000 (6 months * $5000). This covers the ~3 months of build as well as a further 3 months. We will continue hosting it past this point but this gives us an initial baseline, at some point in the future we may apply for a public goods award based on the utility of the platform to cover those costs. Any unused infrastructure budget will roll on to following months. We'll optimize these costs to ensure we're providing value for money to the network but not by sacrificing reliability of the platform. We'll be hosting full open endpoints for community teams and SDF use and of course all of this work will be open source including full deployment documentation for those looking to self host.
Thank you for reading these amendments and if you have any additional queries or requests we'll be happy to accommodate those throughout the build - ultimately we're a partner to this network and part of it. We view this project as essential infrastructure beyond the contents of this submission that both community builders and SDF will securely rely on.
_ _
CTX.com has provided real-time pricing data since 2020, serving millions of low-latency queries across hundreds of asset pairs with greater than 99.99 percent uptime. Our infrastructure currently powers wallets with over a million downloads, payment service providers, and DeFi applications. The service has operated reliably and free of charge for more than five years.
This project upgrades that proven system into a comprehensive, enterprise-ready pricing infrastructure purpose-built for the Stellar ecosystem. The new platform will deliver real-time and historical pricing across all Stellar assets, including both Classic and Soroban tokens. Assets will be pairable against other Stellar assets, BTC, and more than 100 fiat currencies, with USD serving as the normalization anchor.
In addition to the 16 real-time data sources currently integrated, this expansion will include full indexing of Stellar DEX trades, Soroban AMM and DEX activity such as Soroswap, Aquarius, SDEX, and Blend, and integration of all relevant oracle networks. The system is designed with modular ingestion pipelines, enabling the addition of new centralized exchanges, decentralized markets, and oracle providers without changes to core aggregation logic.
All components will be fully open source. Pricing data will be available via a public API. Free rate-limited API access will be provided, with unlimited API keys available at no cost to SDF and SCF-supported teams. Comprehensive documentation, container images, and deployment scripts will be published to support self-hosted deployments.
Please refer to the accompanying technical architecture document for detailed specifications and implementation design.
$138.0K
CTX Rates has been operating in production for over five years, delivering real-time pricing across hundreds of asset pairs and serving millions of low-latency queries. The system currently ingests data from 16 independent sources, applies outlier filtering, supports source weighting, and generates composite cross-asset rates through deterministic aggregation logic.
The infrastructure has powered pricing for tens of millions of transactions and has been relied upon consistently by wallets, payment processors, and DeFi applications. It has maintained greater than 99.99 percent uptime throughout this period.
This project builds on a mature, battle-tested system with demonstrated reliability, operational stability, and real-world usage across multiple production integrations.
Phase 1: Discovery, Architecture Finalization and Environment Setup (Week 1)
Review with stakeholders any suggested modifications to this plan, explore their use cases to ensure end design is suitable.
Finalize system architecture and configuration model
Define aggregation thresholds and default parameters
Provision development and staging environments
Establish CI/CD pipelines
Define monitoring and logging baselines
Deliverable: Confirmed architecture specification and operational development environment.
Budget: 3 developers, 40 hours/week, 1 week, $90/hour - $10800
Phase 2: Core Ingestion and Normalization (Weeks 2–3)
Implement Stellar Classic DEX ingestion
Implement Soroban event indexing for AMMs and DEXs
Integrate initial centralized exchange connectors
Integrate FX anchor feeds
Build canonical normalization pipeline
Deliverable: Unified ingestion layer producing validated, normalized trade records.
Budget: 3 developers, 40 hours/week, 2 weeks, $90/hour - $21600
Phase 3: Aggregation Engine and Spot Pricing (Weeks 4–5)
Implement rolling-window VWAP and TWAP logic
Implement liquidity eligibility rules
Implement triangulation path selection
Implement divergence and outlier filtering
Expose internal spot pricing service
Deliverable: Deterministic spot pricing engine with fallback logic.
Budget: 3 developers, 40 hours/week, 2 weeks, $90/hour - $21600
Phase 4: Historical Aggregation and OHLC (Week 6)
Implement time-bucketed aggregation
Generate OHLC data across required granularities
Implement retention and compaction policies
Validate 1 hour and above indefinite retention
Deliverable: Historical pricing with compliant timeframes.
Budget: 3 developers, 40 hours/week, 1 week, $90/hour - $10800
Phase 5: API Layer and Query Services (Week 7)
Implement REST endpoints
Implement batch query support
Implement rate limiting and access control
Implement optional streaming support
Integrate caching layer
Deliverable: Public API endpoints meeting latency targets in staging.
Budget: 3 developers, 40 hours/week, 1 week, $90/hour - $10800
Phase 6: Infrastructure Hardening and SLA Validation (Weeks 8–9)
Load testing and performance tuning
p95 and p99 latency validation
Multi-zone deployment configuration
Backup and recovery testing
Incident detection and alert validation
Deliverable: Production-grade deployment validated against SLA requirements.
Budget: 3 developers, 40 hours/week, 2 weeks, $90/hour - $21600
Phase 7: Finalization (Week 10)
Final documentation and API reference publication
Public open-source repository release
Production deployment
Deliverable: Fully operational public API with open-source codebase and documentation.
Budget: 3 developers, 40 hours/week, 1 week, $90/hour - $10800
We hold AWS certifications & have enterprise software and infrastructure experience including the infrastructure for a high availability VoIP solution.

No other submissions.