AI + DeFi
Topaz for AI agents
Bring AI to your Topaz workflow. Talk to the hosted agent for quotes, liquidity and voting strategies, or give your own agent the official protocol skill.
Topaz Agent
Your on-chain DeFi strategist
A non-custodial DeFi strategist for BNB Chain, hosted at agents.topazdex.com. It is purpose-built for the Topaz protocol. It uses multi-step AI reasoning with domain-specific playbooks to analyze pools, recommend positions, optimize votes, plan bribe campaigns, and build wallet-ready transactions — all without ever holding your keys.
Non-custodial by design. The agent builds calldata; your wallet signs. Registered on-chain as Agent ID 113284 via ERC-8004.
Swap quotes & routing
Best-price routing across v2 volatile, v2 stable, and v3 concentrated liquidity pools with slippage modeling.
LP recommendations
Risk-scored liquidity position suggestions with full calldata building for both v2 LP and v3 CL positions.
Vote optimization
Gauge vote allocation strategies — maximize yield, support protocol health, or find a balanced mix.
Bribe campaigns
Plan and build incentive deposit transactions with budget scheduling and projected APR impact.
Reward claims
Build claim transactions for emissions, trading fees, bribes, and rebase across all your veNFTs and LPs.
Strategy reports
Comprehensive weekly epoch reports covering protocol state, top pools, voting opportunities, and position health.
Integration Protocols
Four ways to connect
Whether you're a user chatting in natural language, a developer calling REST endpoints, or an autonomous agent connecting via MCP — the Topaz Agent speaks your protocol.
Chat
Web & TelegramNatural language interface with multi-step AI reasoning. Ask questions, get recommendations, build transactions.
https://agents.topazdex.comMCP
AI AgentsProtocol tools via Model Context Protocol (JSON-RPC 2.0). Connect your MCP-compatible agent using this endpoint.
https://agents.topazdex.com/mcpOpenAPI
DevelopersStandard REST API with full spec. Works with any HTTP client, SDK, or automation platform.
https://agents.topazdex.com/openapi.jsonAgent-to-Agent
A2A ProtocolA2A discovery card for autonomous agent-to-agent communication and task delegation.
https://agents.topazdex.com/.well-known/agent-card.jsonAgent Skill
Teach your own agent the Topaz protocol
The Topaz Agent Skill is a runtime-neutral package — markdown instructions, protocol references, worked examples, evals, and TypeScript helpers — that gives any AI agent the structured context it needs to act on the Topaz protocol. Install it into Claude Code, Cursor, Hermes, OpenCode, or any agent with a skill system.
- Version
- See current manifest
- License
- MIT
- Chain
- BNB Chain hub (56), Robinhood (4663), Base (8453), Ethereum (1), Arc (5042)
- Repository
- topazdex/agent-skill
Quick Start
Three ways to load Topaz into your agent
Pick the path that matches your runtime. All three point at the same canonical skill.
URL prompt
Best for browser & chat agents
If your agent can fetch URLs, give it this prompt:
Read the official Topaz DEX agent skill at
https://www.topazdex.com/skill.md and follow it
when working with Topaz DEX swaps, pools,
gauges, bribes, veNFTs, rewards, quotes,
routes, or integrations.Git clone
Best for coding & CLI agents
Full package — references, examples, scripts, evals:
git clone https://github.com/topazdex/agent-skill.git <dest>
cd <dest>/scripts
cp .env.example .env
yarn install --immutable
yarn validate && yarn smokeOne-line installer
Auto-detects your skills dir
Detects ~/.claude/skills, ~/.config/opencode/skills, or ~/.hermes/skills and installs there. Falls through to ~/.local/share/topaz-skill if none are found. Prints its chosen destination before cloning.
curl -fsSL https://raw.githubusercontent.com/topazdex/agent-skill/main/install.sh | bash
# or pick your own path
curl -fsSL https://raw.githubusercontent.com/topazdex/agent-skill/main/install.sh \
| bash -s -- ~/some/other/pathEntry Points
Stable URLs you can point any agent at
These endpoints stay stable across versions. Bookmark them, hard-code them in agent configs, ship them in prompts.
Skill file (markdown)
→Stable bootstrap markdown. Point any URL-reading agent here.
www.topazdex.com/skill.mdManifest (JSON)
→Machine-readable metadata: version, install, verify, compatibility.
www.topazdex.com/skill.jsonGitHub repository
→Canonical source of truth — references, examples, scripts, evals.
github.com/topazdex/agent-skillLatest release
→Tagged versions of the skill, packaged for reproducible installs.
releases/latestChangelog
→Every shipped change. Diff this against your installed version.
CHANGELOG.mdDiscovery
→Tiny .well-known endpoint for agent runtimes that auto-discover skills.
.well-known/agent-skill.jsonConsumption Modes
Works with how your agent already loads context
Not every agent has the same skill system. The Topaz skill is designed to plug into all of them.
Read from URL
Best for browser agents, chat agents, and anything with a web-fetch tool. Point it at skill.md and you're done.
https://www.topazdex.com/skill.mdClone the full package
Best for coding agents, CLI agents, and anything that runs scripts. Includes references/, examples/, evals/, and TypeScript helpers.
git clone github.com/topazdex/agent-skillCopy into a native skill system
If your agent has its own skills, rules, memory, or custom-instructions concept, drop SKILL.md into that system and point at the repo for references.
<runtime>/skills/topazUse as project context
For coding agents, add the skill to your project rules or agent memory so Topaz-specific behavior applies while you work in your repo.
.cursor/rules · .claude/CLAUDE.md · etc.Runtime Examples
Conventional locations per runtime
These are examples, not requirements. The canonical source is the GitHub repo — point your agent at whichever path it reads from.
| Runtime | Conventional location |
|---|---|
| Claude Code (user-wide) | ~/.claude/skills/topaz |
| Claude Code (project-local) | <your-project>/.claude/skills/topaz |
| Hermes | ~/.hermes/skills/defi/topaz |
| OpenCode | ~/.config/opencode/skills/topaz |
| Codex / Cursor / Continue / Aider / generic | any path — point the agent at it explicitly |
| Browser-only / URL-loading agents | read https://www.topazdex.com/skill.md directly |
Generic URL-reading agents
Load https://www.topazdex.com/skill.md and use it as your operating guide for Topaz DEX tasks.Git-capable coding agents
Clone https://github.com/topazdex/agent-skill and follow SKILL.md. Use references, examples, and scripts from the cloned repo when available.Claude Code
Read https://www.topazdex.com/skill.md. If more detail is needed, clone https://github.com/topazdex/agent-skill into ~/.claude/skills/topaz (or .claude/skills/topaz for project-local) and follow SKILL.md as guidance for Topaz DEX tasks.OpenAI Codex / generic code agents
Use https://www.topazdex.com/skill.md as the Topaz DEX protocol guide. For full references and scripts, clone https://github.com/topazdex/agent-skill.Skill Capabilities
What agents can do with the skill
Workflows for all five networks are documented. Quote and swap-batch helpers take any of the five chain IDs, and the remaining helpers target BNB Chain; each helper's reference lists its networks.
Swaps & quotes
v2 volatile, v2 stable, v3 concentrated, and mixed routes with slippage modeling.
Built calldata
Wallet-ready swap and LP transactions. Never broadcast unless the user explicitly asks.
Liquidity positions
v2 LP and Slipstream v3 NFT positions — open, close, inspect, rebalance plans.
Gauges & emissions
BNB TOPAZ emissions per gauge, gauge state, APR math and weekly epochs. Spoke gauges pay xTOPAZ; check the agent package’s chain support before querying them.
veTOPAZ & voting
Lock duration, voting power, vote allocations, reset, and re-vote flows.
Bribes & rewards
Emissions, trading fees, bribes, rebase — claimable across veNFTs and LPs.
Integrations
Quote widgets, swap calldata for bots, dashboards, and subgraph queries.
Debugging
Diagnose failed quotes, broken routes, stuck approvals, and reverted txs.
Safety
Quote, build, simulate — never broadcast by default
Both the Topaz Agent and the agent skill are safe by default. Funds never move unless the user explicitly asks and signing credentials are configured.
Every output carries one of four labels so the user always knows what just happened:
quoteRead-only price/route preview. Nothing on-chain.
built calldataTransaction bytes ready for the user's wallet. Not broadcast.
approval-neededAllowance or veNFT delegation must happen before the next step.
broadcast tx-hashOnly after the user explicitly authorizes and signing is configured.
- ✓Quote before building transactions
- ✓Simulate or dry-run when possible
- ✓Clearly report slippage assumptions
- ✓Clearly report approval requirements
- ✓Build calldata without broadcasting by default
- ✓No fund movement unless the user explicitly asks
Update & Verify
Keep your skill installation current
Compare versions against /skill.json, pull updates, and verify the full package boots cleanly.
Update
If installed as a git repository:
git -C <dest> pull --ff-only
# or use the wrapper:
bash <dest>/update.shCheck whether an update is available without applying it:
cd <dest>
bash tools/check_update.sh
# exit 0 = up to date, 10 = update availableVerify
Full-repo install — node >= 20, yarn@4 (corepack enable):
cd <dest>/scripts
yarn install --immutable
yarn validate
yarn build
yarn test
yarn smokeyarn smoke is a live read against BSC mainnet — set BSC_RPC_URL in scripts/.env. No PRIVATE_KEY needed for reads.
Versioning
Skill: See current manifest
The repo validator enforces that SKILL.md frontmatter version matches skill.json at every release. Compare your local version against the manifest to decide whether to update.
Report Issues
Stale data? Unsafe behavior?
If you find outdated protocol data, broken helpers, incorrect docs, or an agent behaving unsafely, open an issue. The skill is maintained alongside the protocol — every report ships back into the next release.
Open an issueAgent support and multichain scope
The hosted agent service and installable skill are maintained separately from this interface. The published skill covers all five networks and documents xTOPAZ entry and redemption, bridging and spoke voting. Its quote and swap-batch helpers accept all five chain IDs; the remaining helpers and command-line tools target BNB Chain. Check the service's current capabilities and deployed network before using a builder; never substitute BNB addresses or data for a planned spoke.
This interface supports xTOPAZ entry, redemption, bridging and spoke voting on the live networks. The hosted agent service and installable skill have their own release capabilities; web support does not activate agent builders. Check the published service manifest and chain-qualified addresses. Live networks · Multichain integration.
Carry chain, account, contract, token address, recipient and NFT or local stake ID in every quote and plan. Keep source submission, source confirmation and destination delivery distinct. A transaction hash is not proof that a bridge completed or a destination action succeeded.