Open Grant Proposal: Filecoin Deal & Retrieval Health Indexer + Explorer

# Open Grant Proposal: Filecoin Deal & Retrieval Health Indexer + Explorer

**Project Name:** Filecoin Deal & Retrieval Health Indexer + Explorer

**Proposal Category:** Developer and data tooling

**Individual or Entity Name:** Individual (Alexander Lavysh)

**Proposer:** vvuwei

**Project Repo(s):** None yet — new project. Prior related work (evidence of capability, not this project's repo): github.com/vvuwei public repos; production indexer/explorer at celenium.io (Celestia network explorer, same team).

**(Optional) Filecoin ecosystem affiliations:** None currently. No prior work relationship with Protocol Labs, Filecoin Foundation, or FFDW.

**(Optional) Technical Sponsor:** None.

**Do you agree to open source all work you do on behalf of this RFP under the MIT/Apache-2 dual-license?:** Yes.

# Project Summary

Filecoin storage deals and retrieval performance are hard to observe in one place today: deal state lives in chain state / Lotus RPC, storage-provider (SP) health and retrieval success rates live in separate ecosystem tools (if anywhere), and there is no lightweight, self-hostable indexer + explorer that surfaces deal lifecycle, sector state, and SP retrieval health together with a queryable API.

This project builds that indexer: a Go service that ingests Filecoin chain state (deals, sectors, SP info) plus retrieval-market signals, stores them in a queryable Postgres/ClickHouse-style backend, and exposes both a REST/GraphQL API and a lightweight explorer UI. The architecture is a direct port of a pattern already shipped in production for Celestia (Celenium, celenium.io) — an indexer holding sub-5s indexing lag under 128MB blocks, plus a documented "lightweight explorer" variant designed to point at any RPC endpoint and produce an immediately usable block/deal explorer.

## Impact

Developers building on Filecoin (storage dapps, SP tooling, data-onboarding pipelines) currently have to write custom chain-state parsing for even basic questions like "which SPs are retrieval-healthy right now" or "what's the deal success/failure rate for sector X." A shared, open-source indexer + API removes that duplicated work and gives the ecosystem a single place to query deal/retrieval health, the same way block explorers became shared infrastructure for every other major chain.

Pain points addressed: fragmented deal/retrieval visibility, no lightweight self-hostable option (existing explorers are either commercial or full-node-heavy), no queryable API for deal/SP health suitable for agents or dashboards to consume programmatically. Risk of not building this: teams keep re-implementing the same chain-parsing logic, and SP health stays opaque to storage clients evaluating who to deal with.

## Outcomes

Final deliverable: an open-source Go indexer service + REST API + lightweight web explorer, deployable via a single binary/RPC-endpoint config (mirroring the "plug-and-play" pattern already built for Celestia), covering:
- Deal lifecycle (proposed → active → terminated/expired) per client and per SP
- Sector state and proof status
- SP-level retrieval success rate and uptime signal, aggregated over rolling windows
- A queryable API (REST, optionally GraphQL) developers and agents can hit directly

Success metrics: indexer keeps pace with chain head within a defined lag budget (target: comparable sub-minute lag, chain-tip-dependent); API used by at least one third-party integrator or dashboard within 60 days of release; explorer deployed publicly and indexable.

## Adoption, Reach, and Growth Strategies

Target audience: Filecoin storage-dapp developers, SP operators wanting to self-report health, and data-onboarding teams evaluating SPs. Initial distribution: publish to Filecoin Slack/dev forums, submit to awesome-filecoin lists, and cross-link from the proposer's existing explorer/indexer audience (celenium.io draws organic Celestia-ecosystem developer traffic). First 10 users: direct outreach to SP operators and storage-dapp teams already active in Filecoin dev channels. First 100: organic via public explorer + API docs once deployed.

## Development Roadmap

**Milestone 1 — Core chain-state indexer (4 weeks, 1 engineer, $15,000)**
- Go indexer ingesting deals, sectors, and SP metadata from Lotus RPC / chain state into a Postgres-backed store
- Backfill + live-follow modes, retry/reorg handling
- Deliverable: indexer running against Filecoin mainnet, deal/sector data queryable via SQL
- Completion date: ~6 weeks after grant start

**Milestone 2 — Retrieval health signal + public API (3 weeks, 1 engineer, $15,000)**
- Retrieval success-rate and SP uptime aggregation pipeline
- REST API (OpenAPI-documented) exposing deals, sectors, SP health, with pagination/filtering
- Deliverable: public API endpoint + docs
- Completion date: ~9 weeks after grant start

**Milestone 3 — Lightweight explorer UI + public deployment (3 weeks, 1 engineer, $20,000)**
- Web explorer UI (reuse of the already-designed "lightweight explorer" pattern: point at any RPC/API endpoint, get a usable UI immediately)
- Public deployment, docs, open-source release under MIT/Apache-2
- Deliverable: live public explorer + full repo release
- Completion date: ~12 weeks after grant start

## Total Budget Requested

| Milestone # | Description | Deliverables | Completion Date | Funding |
|---|---|---|---|---|
| 1 | Core chain-state indexer | Indexer live on mainnet, deal/sector data queryable | Week 6 | $15,000 |
| 2 | Retrieval health + public API | Public REST API + docs | Week 9 | $15,000 |
| 3 | Explorer UI + public release | Live public explorer, open-source repo release | Week 12 | $20,000 |

**Total: $50,000**

## Maintenance and Upgrade Plans

The proposer currently operates and maintains a production Celestia explorer/indexer under an active multi-milestone partnership, including ongoing infra/optimization work (blob storage optimization, indexer performance work) — the same maintenance discipline applies here. Post-grant, the indexer will be maintained as an open-source project with issue triage and dependency/chain-upgrade compatibility updates; sustainability path is ecosystem adoption plus potential follow-on Filecoin Foundation support for expanded scope (e.g., additional SP health signals, historical analytics).

# Team

## Team Members
- Alexander Lavysh (proposer, engineer)

## Team Member LinkedIn Profiles
- Available on request.

## Team Website
N/A (individual proposal).

## Relevant Experience

Proposer builds and operates Celenium (celenium.io), the canonical public block explorer and indexer for the Celestia network, including PayForBlobs/blob/namespace indexing, a validator/governance surface, and cross-chain (IBC, Hyperlane, Blobstream) interaction tracking. The indexer maintains sub-5s indexing lag under 128MB blocks — a scale/latency bar comparable to what a Filecoin deal/sector indexer needs to hold against chain growth. Also designed and shipped Celestial Ids, a production Web3 identity product (8,800+ mints, 5,900+ unique owners, live across 18 chains, integrated into MetaMask Snap and Leap Wallet). Stack: Go, Cosmos SDK internals, Postgres/ClickHouse-style indexing pipelines, REST/GraphQL APIs, Next.js/Nuxt frontends. No prior Filecoin-specific work — this proposal is a deliberate extension of an existing indexer architecture into a new ecosystem, not a from-scratch attempt.

## Team code repositories
- https://github.com/vvuwei (public repos)
- celenium.io (production explorer; primary backend/indexer repos are private under a foundation SoW, available for reviewer walkthrough on request)

# Additional Information

How did you learn about the Open Grants Program? Public devgrants repository / fil.org/grants.

Best email for discussing the grant agreement: xlavysh@gmail.com

Additional context: happy to scope down to a single milestone (e.g., just the core indexer + API, dropping the explorer UI) if $50k is too large an initial ask — flexible on scope-to-budget ratio.

No project repository, files, tests, or implementation entry points are provided. Start by reviewing the three milestones and the proposer's existing Celenium indexer architecture, then clarify the initial scope and repository before work begins. Done would be an open-source Go indexer with the specified API and explorer deliverables, but the proposal leaves the implementation and validation details open.

Written by the indexing model from the issue text.