Skip to content

Open Grant Proposal: toju — crypto-native payment bridge for pay-as-you-go storage on IPFS #2069

@kaf-lamed-beyt

Description

@kaf-lamed-beyt

Project Name: toju

Proposal Category: Developer and data tooling

Individual or Entity Name: Individual

Proposer: kaf-lamed-beyt

Project Repo(s):

(Optional) Filecoin ecosystem affiliations:
FIL RetroPGF grant recipient (2025). Participant in Protocol Labs Developer Guild (PLDG, 2025–2026).

(Optional) Technical Sponsor: N/A

Do you agree to open source all work under MIT/Apache-2 dual-license?: Yes


Project Summary

Decentralized storage infrastructure exists and is mature — the IPFS network provides reliable, content-addressed storage backed by Filecoin. The bottleneck is the payment layer. Developers building on Solana, Filecoin, or Base cannot pay for decentralized storage programmatically from their wallets. Existing IPFS storage services use subscription-based billing with credit card requirements, which means even a developer with a full crypto wallet has to leave the Web3 stack to purchase storage. Every centralized payment step is a drop-off point.

toju is an open source payment bridge that solves this. It provides a crypto-native, pay-per-upload payment layer on top of the IPFS network that works across chains. Users pay with SOL on Solana, USDFC on Filecoin, or USDC on Base — directly from their wallets, no account creation required. The SDK handles cost estimation, on-chain payment, and storage provisioning in three function calls. The storage backend is built on Helia (the modern JS IPFS implementation), making toju storage-provider agnostic and resilient to changes in any single upstream service.

toju also implements the x402 protocol for autonomous agent payments, enabling AI agents to pay for IPFS storage per-operation using EIP-3009 signed authorizations, with no human in the loop. This is the first implementation of x402 for decentralized storage, and connects the emerging agentic economy directly to Filecoin-backed IPFS infrastructure.

We've built the payment infrastructure. This grant funds migrating to a direct Helia-based IPFS backend and getting the product in front of developers.


Impact

Pain points addressed:

The core pain point is payment friction. A Solana or Filecoin developer cannot today write a single line of code that pays for decentralized storage from their wallet without building significant custom infrastructure. Existing IPFS storage services require email signup and credit card billing. This forces developers to either build their own integration from scratch, use centralized storage like AWS S3, or skip persistent storage entirely. toju removes all three failure modes with an npm install.

A second pain point — exposed by the recent sunset of managed IPFS storage services — is backend lock-in. toju's migration to Helia eliminates this: by integrating directly with the IPFS protocol layer, toju is not dependent on any single storage provider's continued operation. Developers who build on toju are building on the protocol, not a service.

For AI agents specifically, the problem is even sharper. Agents that generate artifacts — documents, images, conversation histories — have no way to pay for persistent storage autonomously. They need a human to set up a subscription first, which breaks the autonomous loop. As agents become the primary consumers of infrastructure, this is a growing and underserved gap.

Benefits of getting this right:

Every developer who integrates the toju SDK is storing data on IPFS, backed by Filecoin. We don't need to convince developers to use Filecoin — we make it the path of least resistance. Success looks like: 50+ active developers using the SDK, measurable USDFC transaction volume on Filecoin mainnet, and at least one AI agent framework community with an active toju integration.

Risks of not getting this right:

Without a crypto-native, provider-agnostic payment layer, developers continue defaulting to AWS S3. The Filecoin/IPFS ecosystem remains technically impressive but practically inaccessible to the developers most likely to build with it — crypto-native teams who already have wallets but no credit cards in their development workflow.


Outcomes

The payment infrastructure is complete and live:

  • USDFC payment verification with receipt polling ✅
  • FIL storage renewal (server + SDK + UI) ✅
  • @toju.network/x402 published on npm ✅
  • Reference applications published on YouTube ✅
  • Developer documentation at docs.toju.network

This grant funds two things: migrating the storage backend to direct IPFS via Helia, and growing developer adoption.

Success metrics:

  • Helia-based backend live and handling production uploads
  • 50+ developers installing @toju.network/fil or @toju.network/x402
  • Measurable USDFC transaction volume growth on Filecoin mainnet
  • 3+ agent framework communities with published toju integration examples
  • 10+ GitHub stars/forks on tojunetwork/afara from external contributors

Data Onboarding

All uploads via toju are stored on IPFS. Projected onboarding as developer adoption grows:

  • Month 1: ~1 GB
  • Month 3: ~25 GB
  • Month 6: ~150 GB
  • Month 12: ~1 TB

Adoption, Reach, and Growth Strategies

Target audiences:

  1. Filecoin/USDFC developers — teams building on Filecoin who want programmatic, wallet-based IPFS storage payments. Reached via Filecoin developer Discord, ecosystem events, and the Filecoin developer newsletter.

  2. AI agent builders — developers using ElizaOS, LangChain, CrewAI, and similar frameworks who need persistent storage for agent outputs. Reached via agent developer communities on Discord and GitHub. toju may be the only storage provider an autonomous agent can pay without human intervention today.

  3. Solana developers — teams building NFT platforms, social apps, and content-heavy applications that currently use AWS S3 or Pinata. Reached via Solana developer forums and the npm registry.

Onboarding strategy:

The SDK is the wedge. pnpm install @toju.network/fil gives developers USDFC-paid IPFS storage in three function calls. We will publish technical walkthrough posts for each integration path, targeting developer communities on Discord, Mirror.xyz, and the Filecoin developer newsletter. For the first 10 developers, direct outreach in Filecoin Discord and GitHub discussions. For 100+, the npm install path compounds: every integrated app is a distribution channel.


Development Roadmap

Milestone 1 — Helia-based IPFS backend migration (Month 1, $10,000)

The current storage backend depends on a managed IPFS service. This milestone replaces it with a direct Helia integration — making toju storage-provider agnostic and resilient.

  • Integrate Helia (JS IPFS) as the core storage layer on the server
  • Implement content pinning via the standard IPFS remote pinning API (provider-agnostic — compatible with any compliant pinning service)
  • Migrate existing upload flow: CID computation, chunking, and pinning via Helia replace the managed service SDK calls
  • Verify CID compatibility — all existing CIDs remain valid and accessible
  • Zero downtime migration: new uploads go via Helia while existing content remains accessible through migration period
  • Team: 1 engineer (full-time for milestone)
  • Funding: $10,000

Milestone 2 — Developer onboarding infrastructure (Month 2, $8,000)

  • Quickstart guides for Filecoin/USDFC developers (docs.toju.network)
  • Integration guides for ElizaOS, LangChain, CrewAI published and cross-posted to ecosystem channels
  • Written + video walkthrough for USDFC payment flow
  • Office hours / async support channel for developers integrating toju

Milestone 3 — Ecosystem presence (Month 3, $8,000)

  • Filecoin developer community engagement (Discord, forums, GitHub discussions)
  • Agent developer community engagement (ElizaOS, LangChain communities)
  • Technical blog posts published on Mirror.xyz and the Filecoin newsletter
  • 10+ developers actively integrating toju (tracked via npm downloads + direct outreach)

Milestone 4 — SDK stability & measurement (Month 4, $4,000)

  • Bug fixes and improvements based on real developer integration feedback
  • @toju.network/fil and @toju.network/x402 patch releases
  • USDFC transaction volume report on Filecoin mainnet
  • npm download metrics report across all three packages
  • Roadmap for next phase (additional chains, team accounts, API keys)

Total Budget Requested

Milestone # Description Deliverables Completion Date Funding
1 Helia-based IPFS backend migration Provider-agnostic storage layer, zero-downtime migration Month 1 $10,000
2 Developer onboarding infrastructure Quickstart guides, integration guides, video walkthrough Month 2 $8,000
3 Ecosystem presence Community engagement, blog posts, 10+ active developers Month 3 $8,000
4 SDK stability & measurement Patch releases, metrics report, next phase roadmap Month 4 $4,000
Total $30,000

Maintenance and Upgrade Plans

toju is a live product at toju.network with real users and a public API. All SDK packages are actively maintained on npm. The GitHub repository (tojunetwork/afara) is public under Apache-2.0. Planned expansions: ETH on Base, additional L2s, team accounts, and API keys for enterprise users. Sustainability is funded by a reseller margin on storage purchases, ecosystem grants, and eventual B2B contracts.


Team

Team Members

Team Website

https://toju.network

Relevant Experience

Built toju end-to-end: Solana Anchor escrow program, multi-chain TypeScript SDKs (@toju.network/sol, @toju.network/fil, @toju.network/x402), Express.js backend, React frontend, and storage lifecycle management system. FIL RetroPGF grant recipient. Protocol Labs Developer Guild (PLDG) participant.

toju is the first storage product to implement x402 for autonomous agent payments, and the only pay-per-upload option that accepts USDFC for Filecoin-backed IPFS storage today.

Team code repositories


Additional Information

Learned about the Open Grants program through participation in the Protocol Labs ecosystem (PLDG, FIL RetroPGF).

Contact email: caleb@toju.network

Links:

Metadata

Metadata

Assignees

No one assigned

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions