# ARBuilder - Final Report

**URL:** <https://forum.arbitrum.foundation/t/arbuilder-final-report/30706>\
**Category:** Domain Allocator Offerings (prev Questbook)\
**Created:** [March 28, 2026, 4:30am UTC](https://forum.arbitrum.foundation/t/arbuilder-final-report/30706 "2026-03-28T04:30:18Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![wms2537](https://yyz1.discourse-cdn.com/flex029/user_avatar/forum.arbitrum.foundation/wms2537/32/16368_2.png) [@wms2537](https://forum.arbitrum.foundation/u/wms2537)\
**Post date:** [March 28, 2026, 4:30am UTC](https://forum.arbitrum.foundation/t/arbuilder-final-report/30706/1 "2026-03-28T04:30:19Z")

</div>

# ARBuilder — Final Report

**Project** : ARBuilder — AI-Powered Development Assistant for the Arbitrum Ecosystem  
**Team** : [Quantum3 Labs](https://github.com/Quantum3-Labs)  
**Grant Program** : Arbitrum Foundation Developer Tooling  
**Period** : January – March 2026  
**Milestones Completed** : 5/5

| Link | URL |
| --- | --- |
| GitHub Repository | [GitHub - Quantum3-Labs/ARBuilder: AI-powered code generator for Arbitrum Stylus, SDK bridging, dApps & Orbit chains — 19 MCP tools with RAG · GitHub](https://github.com/Quantum3-Labs/ARBuilder) |
| Hosted Service | [https://arbuilder.app](https://arbuilder.app) |
| YouTube Channel | [https://www.youtube.com/@Arbuilder-app](https://www.youtube.com/@Arbuilder-app) |
| Glama MCP Listing | [ARBuilder by Quantum3-Labs | Glama](https://glama.ai/mcp/servers/Quantum3-Labs/ARBuilder) |
| Tutorial Video | [https://youtu.be/T0bGSYwjMeo](https://youtu.be/T0bGSYwjMeo) |

* * *

## 1. Overview

ARBuilder is an open-source MCP (Model Context Protocol) server that provides AI-powered code generation for the Arbitrum ecosystem. It transforms natural language prompts into production-ready code for:

- **Stylus smart contracts** (Rust/WASM) — SDK 0.10.0
- **Cross-chain bridging** (ETH/ERC20 across L1, L2, L3)
- **Full-stack dApps** (NestJS + Next.js + The Graph + Chainlink)
- **Orbit chain deployment** (config, rollup, validators, node setup)

The tool uses a RAG (Retrieval-Augmented Generation) pipeline with hybrid BM25 + vector search over 40+ curated Arbitrum repositories, ensuring generated code follows current SDK patterns rather than relying on outdated LLM training data.

**19 MCP tools** work inside Cursor, VS Code, and Claude Desktop — or via the hosted playground at [arbuilder.app](https://arbuilder.app).

* * *

## 2. Deliverables & Milestone Completion

All 5 milestones have been completed and merged to the main branch.

| # | Milestone | Description | Status | PRs |
| --- | --- | --- | --- | --- |
| M1 | Proposal + PoC Improvement | Updated proposal, improved PoC, documentation | Complete | — |
| M2 | Stylus + SDK RAG | Stylus code generation (6 tools), Arbitrum SDK bridging (3 tools), MCP integration, tutorial video | Complete | #4, #5, #6, #7 |
| M3 | Full dApp Builder | Backend, frontend, indexer, oracle, orchestration (5 tools), hybrid RAG, ABI extraction | Complete | #9, #10, #11, #12, #13 |
| M4 | Orbit Integration | Chain config, rollup deployment, validator setup, Q&A, orchestration (5 tools) | Complete | #14 |
| M5 | Final Report + Metrics | ARBuilder v1.0, final report, KPIs | Complete | #15, #16, #17, #19 |

### Key Technical Deliverables

| Component | Details |
| --- | --- |
| MCP Tools | 19 tools across 4 modules (M1: 6, M2: 3, M3: 5, M4: 5) |
| Templates | 7 Stylus contract templates + 9 Orbit deployment templates |
| RAG Pipeline | Hybrid BM25 + vector search (BGE-M3, 1024 dimensions) + cross-encoder reranking |
| Code Validation | Docker-based `cargo check` with up to 3 auto-fix attempts, 6-phase fix architecture |
| Hosted Service | Cloudflare Workers (Workers AI, Vectorize, D1, KV, Queue) |
| Test Suite | 451 unit/integration tests including TypeScript compilation and bash syntax validation |
| Ingestion Pipeline | Worker-native cron (every 6 hours) — scrape, chunk, embed, upsert |

* * *

## 3. Developer Adoption & Usage Metrics

### 3.1 Platform Usage

Data source: Cloudflare D1 `usage_logs` (MCP endpoint)

| Metric | Value |
| --- | --- |
| **Total code generation requests** | 1,921 |
| **Total LLM tokens consumed** | 5,340,565 |
| **Overall success rate** | 99.7% (1,915/1,921) |
| **Average latency** | 15.6 seconds |
| **Peak daily usage** | 333 requests (March 5, 2026) |
| **Active tracking period** | Feb 24 – Mar 8, 2026 |

### 3.2 Requests by Tool

| Tool | Requests | Category |
| --- | --- | --- |
| generate\_stylus\_code | 488 | M1: Stylus |
| generate\_bridge\_code | 327 | M2: SDK |
| generate\_messaging\_code | 183 | M2: SDK |
| ask\_bridging | 183 | M2: SDK |
| orchestrate\_dapp | 94 | M3: dApp |
| generate\_indexer | 84 | M3: dApp |
| generate\_tests | 69 | M1: Stylus |
| ask\_stylus | 68 | M1: Stylus |
| generate\_frontend | 60 | M3: dApp |
| generate\_backend | 57 | M3: dApp |
| get\_workflow | 51 | M1: Stylus |
| ask\_orbit | 44 | M4: Orbit |
| generate\_orbit\_deployment | 42 | M4: Orbit |
| get\_stylus\_context | 38 | M1: Stylus |
| generate\_oracle | 35 | M3: dApp |
| generate\_validator\_setup | 33 | M4: Orbit |
| generate\_orbit\_config | 33 | M4: Orbit |
| orchestrate\_orbit | 32 | M4: Orbit |

### 3.3 Developer Reach

| Channel | Details |
| --- | --- |
| Registered users (arbuilder.app) | 5 accounts with API keys |
| External developers with mainnet deployments | 5 unique developers |
| GitHub maintainers | 4 (wms2537, DiegoFloresWenHao, macoloye, ngjupeng) |
| GitHub stars | 21 |
| YouTube videos | 3 tutorials published |
| Glama quality score | AAA (highest tier) |

* * *

## 4. Mainnet Deployments — Case Studies

Five projects built with ARBuilder have been independently deployed to **Arbitrum One mainnet** by external developers:

| # | Project | Developer | Contract Address | Description |
| --- | --- | --- | --- | --- |
| 1 | [Arbitrum TipJar](https://github.com/jaysun15/arbitrum_tipjar) | jaysun15 | `0xba39c706dc01a10c974d5dc355e0725def232e2e` | On-chain tipping |
| 2 | [ArbiDrop](https://github.com/Jalaa532/ArbiDrop) | Jalaa532 | `0x27ccb5c8da8ae0e2c8b349a0a7d14627429c9569` | Token airdrop distribution |
| 3 | [ArbTokenLaunch](https://github.com/benedicther/ArbTokenLaunch) | benedicther | `0x1Da2ce71c789E43775c374920b345998B1F72abC` | Token launch platform |
| 4 | [Credential Revocation Registry](https://github.com/ajeelll/credential_revocation_registry) | ajeelll | `0x5c31858bd9769cfc29dc36f0cb40bc90cdff321a` | Verifiable credential registry |
| 5 | [DisasterReliefLedger](https://github.com/samdasak/DisasterReliefLedger) | samdasak | `0xff5a49998772f123c26333fa3a033bdb9cd1fdc7` | Disaster relief fund tracking |

**Testnet projects** (Arbitrum Sepolia):

1. [test-erc20-stylus](https://github.com/wms2537/test-erc20-stylus) — ERC20 token on Stylus SDK 0.10.0
2. [arcadedao-orbit](https://github.com/wms2537/arcadedao-orbit) — Orbit chain deployment
3. [prediction-market-dapp](https://github.com/wms2537/prediction-market-dapp) — Full-stack prediction market

* * *

## 5. KPI Achievement

| KPI | Target | Achieved | Evidence |
| --- | --- | --- | --- |
| Unique developers | ≥10 | **12** | Registered users + GitHub contributors + mainnet deployers (deduplicated) |
| Code generation requests | ≥100 | **1,921** (19.2x target) | Cloudflare D1 `usage_logs` — tracked Feb 24 – Mar 8, 2026 ([schema](https://github.com/Quantum3-Labs/ARBuilder/blob/main/apps/web/migrations/0001_initial_schema.sql)) |
| GitHub stars | ≥20 | **21** | [GitHub](https://github.com/Quantum3-Labs/ARBuilder) |
| Community contributions | ≥5 | **8 issues, 18 merged PRs** | [Issues](https://github.com/Quantum3-Labs/ARBuilder/issues?q=is%3Aissue) / [PRs](https://github.com/Quantum3-Labs/ARBuilder/pulls?q=is%3Apr+is%3Amerged) |
| Case studies | ≥3 | **5** | See Section 4 — all with on-chain contract addresses |
| Projects deployed to mainnet | ≥5 | **5** | See Section 4 — verified on Arbitrum One |

* * *

## 6. Community Engagement & Distribution

### 6.1 Open Source

| Item | Details |
| --- | --- |
| License | MIT |
| README | Comprehensive with architecture, quick start, 19-tool reference |
| CONTRIBUTING.md | Development guide, code style, PR process |
| Issue templates | Bug report + feature request |
| Good first issues | 7 open issues labeled for new contributors |

### 6.2 Directory Listings

| Platform | Status |
| --- | --- |
| [Glama](https://glama.ai/mcp/servers/Quantum3-Labs/ARBuilder) | Listed — AAA score |
| [awesome-mcp-servers](https://github.com/punkpeye/awesome-mcp-servers) | Listed |
| [awesome-stylus](https://github.com/nicbaz/awesome-stylus) | PR #26 submitted |
| [mcp.so](https://mcp.so) | Issue #788 submitted |

### 6.3 Content & Tutorials

| Content | Link |
| --- | --- |
| Setup & usage walkthrough | [YouTube](https://youtu.be/T0bGSYwjMeo) |
| Prediction market dApp demo | [YouTube](https://youtu.be/61L9n3lBgxM) |
| L3 Orbit chain demo | [YouTube](https://youtu.be/7i3fctk2ND0) |

### 6.4 Website & SEO

[arbuilder.app](https://arbuilder.app) includes:

- Landing page with all 4 milestone feature sections
- Interactive playground with all 19 tools
- Transparency page (data sources, pipeline status)
- SEO: JSON-LD structured data, Open Graph, Twitter cards, robots.txt, sitemap.xml
- LLM discovery endpoint ([/llms.txt](https://arbuilder.app/llms.txt)) for AI search engines

* * *

## 7. Technical Architecture

### RAG Pipeline

```auto

Query → BM25 + Vector Search (BGE-M3) → RRF Fusion → Cross-Encoder Reranking → LLM Generation

```

1. **Hybrid search** over 40+ curated Arbitrum repos and docs
2. **Cross-encoder reranking** for precision
3. **65 Stylus compilation rules** in system prompts
4. **Compile verification** via Docker `cargo check` with 3 auto-fix attempts

### Code Quality Pipeline

6-phase fix architecture (Python + TypeScript, kept in sync):

1. Structural corrections (attributes, extern crate)
2. Import resolution (alloc, alloy\_sol\_types)
3. Storage accessor patterns (getter/setter)
4. API migration (SDK 0.9.x → 0.10.0)
5. Type corrections (B256, U256, as\_u64)
6. Output sanitization (garbled LLM output)

### Deployment

| Mode | Stack |
| --- | --- |
| Self-hosted | Python MCP server via stdio → Cursor / VS Code / Claude Desktop |
| Hosted | Cloudflare Workers, Workers AI, Vectorize, D1, KV, Queue |

* * *

## 8. Key Learnings

1. **RAG outperforms fine-tuning for fast-moving SDKs.** Stylus SDK went from 0.6 → 0.10.0 in months. RAG allows knowledge base updates in hours, not weeks of retraining.
2. **Compile verification is non-negotiable.** The `cargo check` + auto-fix loop catches errors that static analysis cannot, especially Stylus-specific patterns like `self.vm()` migration.
3. **Hybrid search matters.** BM25 alone misses semantic similarity; vector search alone misses exact API signatures. The combination with cross-encoder reranking significantly improves code quality.
4. **Template + LLM hybrid works best.** Templates provide reliable scaffolding (100% success rate); LLM fills in custom logic. This gives consistency for common patterns while maintaining flexibility.
5. **Testing catches real bugs.** Our TypeScript compilation tests caught a template substitution ordering bug where `CHAIN_ID_PLACEHOLDER` was being replaced before `PARENT_CHAIN_ID_PLACEHOLDER`, corrupting parent chain IDs in generated code.

* * *

## 9. Repository Statistics

| Metric | Value |
| --- | --- |
| Total commits | 218 |
| Lines of code (Python + TypeScript) | 56,700+ |
| Test cases | 451 |
| Pull requests merged | 18 |
| Issues created | 8 (7 open “good first issue”) |
| Data sources in knowledge base | 40+ |

* * *

## 10. Future Plans

- Continue maintaining the knowledge base as Stylus SDK evolves
- Respond to community issues and contributions via the 7 open “good first issue” items
- Monitor and update directory listings (awesome-mcp-servers, awesome-stylus, mcp.so)
- Explore additional Arbitrum ecosystem integrations based on community feedback
- Track mainnet deployments from new users

* * *

_ARBuilder is open-source under the MIT license. Built by [Quantum3 Labs](https://github.com/Quantum3-Labs) for the Arbitrum ecosystem._

---

<div class="post-metadata">

**Author:** ![Arb\_Junior](https://yyz1.discourse-cdn.com/flex029/user_avatar/forum.arbitrum.foundation/arb_junior/32/16472_2.png) [@Arb\_Junior](https://forum.arbitrum.foundation/u/Arb_Junior)\
**Post date:** [August 21, 2026, 11:54am UTC](https://forum.arbitrum.foundation/t/arbuilder-final-report/30706/2 "2026-08-21T11:54:55Z")

</div>

Thank you for the final report @wms2537

You shipped a working open-source tool that generates Stylus, bridging, dApp, and Orbit code. You hit every milestone, posted strong metrics, and helped five independent projects reach mainnet. The focus on real compile checks and up-to-date knowledge is especially valuable. Well done.

From a governance standpoint (grant oversight, accountability, value-for-money, transparency, community impact, and long-term sustainability within the Arbitrum Foundation Developer Tooling program), this is a strong, well-structured final report. It demonstrates clear milestone delivery, measurable outcomes that substantially exceed targets, verifiable on-chain evidence, and responsible open-source practices.

**Key Governance Strengths:**

• **Full milestone completion with traceability** : All 5 milestones completed and linked to specific PRs. This provides clear auditability of progress against the original grant scope.

• **KPI over-delivery with evidence** : Targets were met or significantly exceeded (e.g., 1,921 code generation requests vs. ≥100; 5 mainnet deployments vs. ≥5; 12 unique developers vs. ≥10). Metrics are backed by concrete sources (Cloudflare D1 logs, GitHub, on-chain addresses).

• **Transparency & verifiability** : Public GitHub repo, hosted service, YouTube tutorials, Glama listing, and five independent mainnet contract addresses allow independent verification.

• **Open-source hygiene** : MIT license, CONTRIBUTING.md, issue templates, good-first-issues, and directory listings show intent to foster community ownership rather than a closed tool.

• **Technical rigor relevant to ecosystem health** : RAG pipeline, compile verification with auto-fix, hybrid search, and SDK version awareness address real pain points (rapid Stylus SDK evolution). This supports broader developer productivity and reduces friction for Arbitrum builders.

• **Usage quality signals** : 99.7% success rate, detailed tool-level breakdown, and external mainnet deployments indicate real utility rather than vanity metrics.

**Areas of Constructive Governance Attention:**

• Adoption is still early (21 stars, limited registered users).

• Usage data covers only a short window.

• Long-term maintenance plan needs more detail.

**Overall:** The team delivered useful tooling with clear impact. Good value for the grant.

**Opinion:**

This grant produced clear results. The over-performance on KPIs and independent mainnet use give solid proof of impact. The technical approach is practical. Continued community growth and maintenance will decide how lasting the tool becomes. At this stage, the delivery looks strong.

**Questions:**

1. How will you keep the knowledge base updated after the grant ends?

2. Do you have usage or feedback data after mid-March?

3. What plans do you have to grow contributors and users?

4. Can you track downstream impact (e.g., more mainnet deployments or time saved)?

5. Are there ecosystem partners or docs sites where closer integration would help?

6. How easy would it be for others to maintain the project if needed?