TronLink Skills

Overview

GitHub: https://github.com/TronLink/tronlink-skills

TronLink Wallet Skills is an AI Agent skill set that provides complete TRON blockchain wallet and DeFi functionality through natural language. Designed for Claude Code, Cursor, OpenCode, Codex CLI, and other AI agents.

Key Highlights: - 6 skills, 33 commands covering wallet, token research, market data, swaps, resources, and staking - Zero npm dependencies — uses native Node.js 18+ fetch and crypto, no npm install needed - TRON-specific domain knowledge — dedicated handling of Energy + Bandwidth resource model - Multi-platform support — Claude Code, Cursor, OpenCode, Codex CLI, LangChain/CrewAI - Read-only & safe — all commands are query-only, no private keys or signing involved - MCP server wrapper for structured AI agent integration

Why TronLink Skills?

TRON has a fundamentally different fee model than EVM chains. Instead of unified gas, TRON uses Energy (for smart contracts) and Bandwidth (for all transactions). No existing AI agent skill properly covers:

• TRON's unique resource model and cost optimization strategies
• Stake 2.0 (freezing TRX to obtain resources and earn rewards)
• Super Representative (SR) voting mechanics
• DEX aggregation across SunSwap V2/V3 and Sun.io
• The 14-day unfreezing wait period and its implications

TronLink Skills fills this gap with deep TRON-specific domain knowledge.

Architecture

Natural Language Input
         |
         v
AI Agent (Claude Code / Cursor / OpenCode / Custom)
         |
         v
tron_api.mjs (Node.js 18+, native fetch, zero dependencies)
    ├── Zero npm dependencies
    ├── TronGrid HTTP API (public or with API key)
    └── Tronscan API for token metadata
         |
         v
Structured JSON → Agent interprets → Natural language response

The 6 Skills

1. tron-wallet (6 commands)

Wallet queries and account information.

Features: Handles both Base58Check (T...) and hex address formats, supports known token symbols, auto-converts decimals.

When NOT to use: Sending TRX/tokens — these are read-only; use the signer SDK or MCP Server TronLink. For deep token-level analytics (rug-pull / liquidity locks), prefer tron-token.

2. tron-token (7 commands)

Token research and security analysis.

Features: Detects rug pulls, analyzes holder concentration, checks liquidity locks.

When NOT to use: Per-account balances or transaction history — that's tron-wallet. Real-time price/volume — that's tron-market.

3. tron-market (8 commands)

Real-time market data and whale monitoring.

Features: Multi-DEX aggregation, smart money signal detection, K-line analysis.

When NOT to use: Quotes or routes for swapping right now — that's tron-swap (which factors in slippage). Static token metadata — tron-token.

4. tron-swap (3 commands)

DEX swap quotes and route optimization.

Features: Aggregates liquidity from multiple sources, estimates Energy cost, handles multi-hop routes.

When NOT to use: Executing the swap — quotes are read-only; the swap itself goes through MCP Server TronLink (tl_chain_swap_v3) or the signer SDK. Historical trade data — tron-market.

5. tron-resource (6 commands)

Energy & Bandwidth management — TRON-specific.

Features: Decision tree logic for cost optimization, tracks daily free bandwidth, calculates TRX burn equivalent.

When NOT to use: Actually freezing TRX to acquire Energy/Bandwidth — that's a Remote Write; use the signer SDK / MCP Server. SR voting strategy after freezing — see tron-staking.

6. tron-staking (3 commands)

Stake 2.0 queries and SR information.

Features: Stake 2.0 status queries, APY calculation, SR commission tracking.

When NOT to use: Performing the freeze/vote/unfreeze actions — those are Remote Writes; route through the signer SDK / MCP Server. Calculating Energy cost per operation — tron-resource.

Skill ↔ MCP Tool Map

scripts/mcp_server.mjs (the wrapper from Method 2) exposes 25 of the 33 commands as MCP tools — every signature, every input field, every output shape is generated from the same tron_api.mjs implementation, so the CLI and the MCP tool are guaranteed equivalent. The 8 CLI-only commands stay reachable through Method 1 (skill prompt) and Method 3 (direct CLI). Use this table when an agent needs to route a user request to a specific tool or when you're inspecting tools/list output.

Totals. 33 CLI commands · 25 MCP tools · 8 CLI-only commands. Every command is read-only — no signing, no broadcast, no fund movement. To execute a transaction, route to MCP Server TronLink (tl_chain_*) or signer SDK (sendTrx, sendTrc20, sign*).

Intent → Skill → Tool Routing

Pick a skill first by the kind of question, then a command by the field the user asked about. Common intents:

If the request implies changing on-chain state (transfer, swap execution, freeze, vote, claim), this skill set is the wrong layer — see the routing notes under each skill's "When NOT to use".

❌ When NOT to route here (negative examples)

Skills are read-only. If the user intent implies a signed / Remote Write action, do not dispatch to a skill — the underlying command will succeed but only as a query/estimate, and the user's actual goal will go unfulfilled. Route to the signer SDK or mcp-server-tronlink instead:

Heuristic. If the user's verb is send / freeze / unfreeze / vote / unvote / approve / swap (execute) / claim / sign / broadcast, the answer never starts in this Skills set. Skills can still precede the write (quote, estimate cost, validate address, check balance) — just don't claim a Skills call finished the user's request.

Recommended Skill Workflows

Balance & Token Check

tron-wallet (check balance) → tron-wallet (list tokens) → tron-resource (check energy status)

Research & Swap Quote

tron-token (search) → tron-market (price/chart) → tron-resource (check energy) → tron-swap (get quote)

Staking Analysis

tron-wallet (check balance) → tron-staking (staking info) → tron-staking (APY estimate) → tron-staking (SR list)

Resource Optimization

tron-resource (check status) → tron-resource (estimate cost) → tron-resource (optimize-cost)

TRON Resource Model Reference

Energy vs. Bandwidth

Cost Examples

Effective as of 2026-05. Energy figures shift with contract upgrades (notably USDT TRC-20) and network parameters; the bandwidth/energy unit prices used to derive the TRX-burned column also change. Treat these as order-of-magnitude reference, not contractual values. Source: TronGrid / Tronscan transaction samples and TRON network parameters. For runtime accuracy, call the tron-resource skill's estimate-energy / estimate-bandwidth commands.

Stake 2.0 Key Facts

• Freeze TRX → Get Energy or Bandwidth → Vote for SR → Earn rewards
• Unfreezing has 14-day wait before withdrawal
• Votes reset if you unfreeze; must re-vote after re-freezing
• 1 frozen TRX ≈ 4.5 Energy/day — dynamic: depends on total network stake; use tron-resource estimate-energy for the live value. Effective figure as of 2026-05.
• Voting rewards claimable every 6 hours

Integration Methods

Method 1: Claude Code (Recommended)

# Clone and use directly
git clone <repo>
cd tronlink-skills
claude   # Auto-discovers SKILL.md files

No npm install needed for read-only operations.

Method 2: MCP Server

# Register as MCP server
claude mcp add tronlink -- node ~/.tronlink-skills/scripts/mcp_server.mjs

# Provides 25 MCP tools callable by Claude Desktop / Claude Code
# (see "Skill ↔ MCP Tool Map" above for the per-command mapping; 8 commands are CLI-only)

Method 3: Manual CLI

# Direct command execution
node scripts/tron_api.mjs wallet-balance --address TAddress...
node scripts/tron_api.mjs token-price --token USDT
node scripts/tron_api.mjs swap-quote --from TRX --to USDT --amount 100

Method 4: Other AI Platforms

Cursor (or Windsurf, same schema)

~/.cursor/mcp.json:

{
  "mcpServers": {
    "tronlink-skills": {
      "command": "node",
      "args": ["/absolute/path/to/tronlink-skills/scripts/mcp_server.mjs"]
    }
  }
}

Codex CLI

# Symlink the skills bundle into the agent's discovery path
ln -s "$(pwd)/tronlink-skills" ~/.agents/skills/tronlink-skills
# Verify discovery
codex skills list | grep tron

OpenCode

~/.config/opencode/config.json:

{
  "plugins": {
    "tronlink-skills": { "path": "/absolute/path/to/tronlink-skills" }
  }
}

LangChain / CrewAI (Python)

from langchain_core.tools import Tool
import subprocess, json

def tron_call(cmd: str) -> dict:
    out = subprocess.check_output(
        ["node", "scripts/tron_api.mjs", *cmd.split()],
        cwd="/absolute/path/to/tronlink-skills",
    )
    return json.loads(out)

tron_wallet_balance = Tool.from_function(
    func=lambda addr: tron_call(f"wallet-balance --address {addr}"),
    name="tron_wallet_balance",
    description="TRX balance and frozen amounts. Address is a TRON Base58 (T...) string.",
)

Quick Setup Script

# Auto-install for all AI environments
bash install.sh

# Clean uninstall
bash uninstall.sh

Configuration

Environment Variables

# Optional: TronGrid API key for higher rate limits
export TRONGRID_API_KEY="your-api-key"

# Optional: Switch network (default: mainnet)
export TRON_NETWORK="mainnet"    # or "shasta" / "nile"

Network Support

Built-In Token Shortcuts

Project Structure

tronlink-skills/
├── README.md                          # Main documentation
├── package.json                       # Node.js manifest
├── install.sh                         # Auto-install for all AI environments
├── uninstall.sh                       # Clean uninstall
│
├── scripts/
│   ├── tron_api.mjs                   # Main CLI (33 commands, zero dependencies)
│   └── mcp_server.mjs                 # MCP protocol server wrapper
│
├── skills/                            # Skill definitions (auto-discovered)
│   ├── tron-wallet/SKILL.md
│   ├── tron-token/SKILL.md
│   ├── tron-market/SKILL.md
│   ├── tron-swap/SKILL.md
│   ├── tron-resource/SKILL.md
│   └── tron-staking/SKILL.md
│
├── docs/
│   ├── claude-integration-guide.md    # 3 integration methods
│   ├── resource-model.md              # Energy & Bandwidth deep dive
│   ├── staking-guide.md               # Stake 2.0 & APY explanation
│   └── integration-guide.sh
│
├── .claude-plugin/                    # Claude Code plugin config
├── .cursor-plugin/                    # Cursor IDE plugin config
├── .opencode/                         # OpenCode config
├── .codex/                            # Codex CLI setup
├── .claude/                           # Pre-configured test commands
├── _meta.json                         # Metadata for skill registries
└── LICENSE                            # MIT

Dependencies

Security Model

Address Format Support

Both formats are supported and auto-normalized across all commands:

Key Design Decisions

1. Zero Dependencies — No npm install required, making it lightweight and instant for AI agents
2. Read-Only & Safe — All commands are queries only, no private keys or signing involved
3. TRON-Specific Domain Knowledge — Dedicated skills for Energy/Bandwidth and Stake 2.0, acknowledging TRON's unique architecture
4. Multi-Format Address Support — Handles both Base58Check and hex formats transparently
5. Token Symbol Resolution — Common tokens have built-in shortcuts; unknown contracts work by address
6. Cost Optimization Recommendations — The optimize-cost command provides personalized strategies
7. MCP Server Wrapper — Provides structured integration for Claude Desktop and modern AI agents

Quick Start

# 1. Clone
git clone <repo>
cd tronlink-skills

# 2. Use with Claude Code (no install needed for reads)
claude
> "What's the TRX balance of TAddress...?"
> "Show me the top trending tokens on TRON"
> "How much Energy do I need to send USDT?"
> "What's the best way to get Energy — freeze, rent, or burn?"

# 3. Or use directly
node scripts/tron_api.mjs wallet-balance --address TAddress...
node scripts/tron_api.mjs token-price --token USDT
node scripts/tron_api.mjs optimize-cost --address TAddress...

Version & License

• Package: tronlink-skills v1.0.1
• License: MIT — SPDX-License-Identifier: MIT
• Changelog / releases: https://github.com/TronLink/tronlink-skills/releases — no GitHub-tagged releases yet for v1.0.x; track changes by commit until the first tag.

Compatibility & migration policy

Skills are at v1.0.x, so standard semver applies — only major bumps may break the public surface.

• Stable contracts (won't change in a minor or patch):
  • The 33 CLI command names and their required / optional flags (tron_api.mjs <command> [...]).
  • The 25 MCP tool names listed in Skill ↔ MCP Tool Map (tron_* form) and their inputSchema keys.
  • Exit codes: 0 success, 1 query error / invalid input, 2 unsupported / unknown command.
  • The Network Read side-effect classification — no command will ever become a Remote Write without a major bump.
• Volatile contracts (may change in a minor):
  • The exact field layout of JSON stdout payloads — new fields can be added in any minor; renames or removals are major. Use a tolerant parser.
  • Built-in token-symbol shortcut list (USDT, USDC, WTRX, …) — symbols may be added in any minor; existing mappings won't be repointed in a minor.
  • Heuristics and thresholds (whale-transfers default cutoff, optimize-cost decision tree weights, etc.).
• Subset relationship. The MCP tool subset (currently 25 of 33) may grow in a minor (a previously CLI-only command exposed as an MCP tool); it will not shrink in a minor.
• Deprecation window. A command / tool marked deprecated continues to work for at least one minor cycle; the runtime prints a STDERR: [DEPRECATED] warning. Removal lands no earlier than the next major.
• Verifying after upgrade. Re-run tron_api.mjs --help and (if using MCP) tools/list to confirm the names you depend on are still present. The MCP serverInfo.version exposed during initialize should match the bumped package.json version.