Skip to content
Agentmart
Submit a tool

stellar-forge

Eval-driven coding harness for building production Stellar dApps.

Skill by rylsherdamz-rgb

IndexedMIT★ 17
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/rylsherdamz-rgb-stellar-forge

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Eval-driven coding harness for building production Stellar dApps. Routes tasks to 6 specialist agents, verifies output with structured evals, steers on failure, e2e-tests with Stellar Agentic Kit hooks, and knowledge-graphs every project. Use when building any Stellar app — tokens, DeFi, NFTs, payment APIs, or full dApps with smart contracts + frontend + backend.

Use cases: frontend, backend, testing, finance

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md and code. Checked Oct 26, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found
Runs as
Node.js package

Network hosts it talks to 13

  • api.github.com · found in code
  • codeload.github.com · found in install command
  • design-tokens.org · found in code
  • fonts.googleapis.com · found in code
  • friendbot.stellar.org · found in code
  • github.com · found in code
  • horizon-testnet.stellar.org · found in code
  • horizon.stellar.org · found in code
  • images.pexels.com · found in code
  • pexels.com · found in code
  • soroban-testnet.stellar.org · found in code
  • stellar-agentic-framework.vercel.app · found in code
  • stellar.expert · found in code

Files it touches

  • ~/.claude/skills/stellar-forge
  • /home/richie/Projects/stellar-forge
  • Your home directory (~)

What's inside

Skill name
stellar-forge
Description
Eval-driven coding harness for building production Stellar dApps. Routes tasks to 6 specialist agents, verifies output with structured evals, steers on failure, e2e-tests with Stellar Agentic Kit hooks, and knowledge-graphs every project. Use when building any Stellar app — tokens, DeFi, NFTs, payment APIs, or full dApps with smart contracts + frontend + backend.
License
MIT
Instructions
176 lines in SKILL.md · view on GitHub

Before installing

  • Ships executable scripts
  • Instructions call the network

Files (200)

  • marketplace.json · .claude-plugin/
  • plugin.json · .claude-plugin/
  • deploy.md · .claude/commands/
  • graphify.md · .claude/commands/
  • scaffold.md · .claude/commands/
  • test-e2e.md · .claude/commands/
  • banner-design · .claude/skills/
  • brand · .claude/skills/
  • brandkit · .claude/skills/
  • design · .claude/skills/
  • design-system · .claude/skills/
  • design-taste-frontend · .claude/skills/
  • design-taste-frontend-v1 · .claude/skills/
  • full-output-enforcement · .claude/skills/
  • gpt-taste · .claude/skills/
  • high-end-visual-design · .claude/skills/
  • image-to-code · .claude/skills/
  • imagegen-frontend-mobile · .claude/skills/
  • imagegen-frontend-web · .claude/skills/
  • industrial-brutalist-ui · .claude/skills/
  • minimalist-ui · .claude/skills/
  • redesign-existing-projects · .claude/skills/
  • slides · .claude/skills/
  • stitch-design-taste · .claude/skills/
  • ui-styling · .claude/skills/
  • ui-ux-pro-max · .claude/skills/
  • .env.example
  • FUNDING.yml · .github/
  • bug-report.md · .github/ISSUE_TEMPLATE/
  • config.yml · .github/ISSUE_TEMPLATE/
  • feature-request.md · .github/ISSUE_TEMPLATE/
  • PULL_REQUEST_TEMPLATE.md · .github/
  • ci.yml · .github/workflows/
  • .gitignore
  • .mcp.json
  • CLAUDE.md
  • CODE_OF_CONDUCT.md
  • CONTRIBUTING.md
  • LICENSE
  • PRODUCT.md
  • README.md
  • SECURITY.md
  • SKILL.md
  • SPEC.md
  • stellar-backend.md · agents/
  • stellar-contracts.md · agents/
  • stellar-frontend.md · agents/
  • stellar-ops.md · agents/
  • stellar-payments.md · agents/
  • stellar-zk.md · agents/
  • .gitignore · contracts/
  • Cargo.lock · contracts/
  • Cargo.toml · contracts/
  • Cargo.toml · contracts/bounty-escrow/
  • lib.rs · contracts/bounty-escrow/src/
  • test.rs · contracts/bounty-escrow/src/
  • approve_before_submit_rejected.1.json · contracts/bounty-escrow/test_snapshots/test/
  • cancel_claimed_bounty_rejected.1.json · contracts/bounty-escrow/test_snapshots/test/
  • cancel_open_bounty_refunds_creator.1.json · contracts/bounty-escrow/test_snapshots/test/
  • claim_after_deadline_rejected.1.json · contracts/bounty-escrow/test_snapshots/test/
  • …and 140 more

SKILL.md (first lines)

# Stellar Agentic Framework

Build production Stellar dApps with an eval-driven, multi-agent coding harness.

## When to Use

- Building a Stellar smart contract (token, DeFi, NFT, zk)
- Scaffolding a full dApp with wallet connect + contract interaction
- Setting up x402 payment middleware for a paid API
- Creating a Stellar frontend with Stellar Wallets Kit
- Writing backend services that query Stellar chain data
- Need eval-driven quality gates for Stellar development

## How It Works

The harness routes your request to 6 specialist agents running in parallel, verifies every output against structured evals, steers on failure (max 3 retries), and produces an eval report.

```
User Request → Route to Agents → Parallel Build → Verify with Evals → Report
                                  ┌──────────────┐
                                  │ @contracts   │
                                  │ @frontend    │
                                  │ @backend     │
                                  │ @payments    │
                                  │ @ops         │
                                  │ @zk          │
                                  └──────────────┘
```

## Setup

### Install the Framework

```bash
npx create-stellar-agentic my-dapp
cd my-dapp
npm install
```

This scaffolds a full project with contracts, frontend, and backend templates.

### Or Install as a Skill Only

```bash
npx create-stellar-agentic --skill-only .
```

Or from the registry:

```bash
npx skills add rylsherdamz-rgb/stellar-forge
```

## Workflow

### Phase 1 — Intent Capture
Describe what you're building. The harness maps keywords to eval templates and writes a decision record.

| Request | Evals Used |
|---------|-----------|
| "Build a token contract" | contract + asset |
| "DeFi lending platform" | contract + backend |
| "Paid API with x402" | agentic-payments |
| "NFT marketplace" | contract + frontend |
| "Full dApp" | all evals |

### Phase 2 — Plan & Spawn Agents
The kernel first plans the work graph (nodes, edges, waves, file ownership — written to `data/graphs/<task>.json`), then routes sub-tasks to specialist agents **in parallel waves** — every independent agent of a wave is spawned in one batch. Each agent receives your intent, eval criteria, owned paths, and interface contract.

See `references/agents.md` for the full agent registry and `references/work-graph.md` for the parallel execution protocol.

### Phase 3 — Verify & Steer
After agents complete, evals check every output. Failed evals trigger corrective retries (max 3).

Evals cover: contract compilation, TypeScript types, wallet connect, contract read/write, server startup, x402 payment flow, and graph health.

See `references/evals.md` for full criteria.

### Phase 4 — E2E Test
1. Start local Stellar network
2. Deploy contracts
3. Run Playwright e2e tests
4. Run Stellar Agentic Kit payment tests

### Phase 5 — Report
```
STELLAR CODING HARNESS :: EVAL REPORT
Project: <name>
CONTRACT EVALS:     X/Y passed
FRONTEND EVALS:     X/Y passed
BACKEND EVALS:      X/Y passed
PAYMENT EVALS:      X/Y passed
E2E EVALS:          X/Y passed
Overall:            SHIP IT / NEEDS WORK
```

## Agentic Kit — Blockchain Data Queries

All blockchain queries use `useStellarData()` — never raw RPC or curl.

```tsx
import { useWallet } from "@/providers/wallet-provider";
import { useContract } from "@/hooks/use-contract";

function MyComponent({ contractId }: { contractId: string }) {
  const { address, getBalances } = useWallet();
  const { read, write } = useContract(contractId);
  // read() = simulation only, no fees
  // write() = full sign + submit + confirm
}
```

For the complete hook API, see `references/agentic-kit.md`.

## Examples

### Input
"Create a token contract with a React frontend to display balances"

From rylsherdamz-rgb/stellar-forge, MIT.

More from rylsherdamz-rgb/stellar-forge

Signals

Stars
17
Forks
2
Open issues
8
Last commit
Sep 9, 2026 (1 month ago)
Last release
Jul 30, 2026 (3 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from rylsherdamz-rgb

Cite this page

stellar-forge skill. Agent Mart. https://agentmart.dev/skills/rylsherdamz-rgb-stellar-forge (last updated 2026-10-24)