Skip to content

sashazykov/dexranger-skill

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 

Repository files navigation

DEX Ranger Skill

Check any crypto token for rug pull risk using DEX Ranger. Returns trust scores, warnings, red flags, and detailed token info for Ethereum, Solana, BSC, and TON tokens.

No API key required. Only needs curl and python3 (standard library only).

This skill follows the open Agent Skills standard and works with 20+ AI agents.

Installation

Universal installer (skills.sh)

The fastest way to install across any supported agent. Requires Node.js.

# Install to a specific agent
npx skills add sashazykov/dexranger-skill -a claude-code
npx skills add sashazykov/dexranger-skill -a codex
npx skills add sashazykov/dexranger-skill -a gemini-cli
npx skills add sashazykov/dexranger-skill -a cursor
npx skills add sashazykov/dexranger-skill -a opencode

# Install to all detected agents at once
npx skills add sashazykov/dexranger-skill --all

See skills.sh for the full list of supported agents.


Claude Code

Personal skill (all projects):

mkdir -p ~/.claude/skills/dexranger
cp SKILL.md ~/.claude/skills/dexranger/SKILL.md
cp -r scripts ~/.claude/skills/dexranger/scripts

Project skill (current project only):

mkdir -p .claude/skills/dexranger
cp SKILL.md .claude/skills/dexranger/SKILL.md
cp -r scripts .claude/skills/dexranger/scripts

Once installed, Claude will automatically use the skill when you ask about token safety, or invoke it directly with /dexranger.

Claude.ai (Web & Desktop App)

  1. Go to Settings > Capabilities
  2. Click Upload skill
  3. Upload the dexranger.skill file from this repository

To generate the .skill file yourself:

zip -r dexranger.skill SKILL.md scripts/

Claude API

Upload the .skill file via the Skills API.


OpenAI Codex

Codex supports the same Agent Skills format. Place the skill in your project or home directory:

Project-level:

mkdir -p .codex/skills/dexranger
cp SKILL.md .codex/skills/dexranger/SKILL.md
cp -r scripts .codex/skills/dexranger/scripts

User-level:

mkdir -p ~/.codex/skills/dexranger
cp SKILL.md ~/.codex/skills/dexranger/SKILL.md
cp -r scripts ~/.codex/skills/dexranger/scripts

Codex also discovers skills from .claude/skills/ and .agents/skills/ directories.

See Codex Agent Skills docs.


Gemini CLI

Project-level:

mkdir -p .gemini/skills/dexranger
cp SKILL.md .gemini/skills/dexranger/SKILL.md
cp -r scripts .gemini/skills/dexranger/scripts

User-level:

mkdir -p ~/.gemini/skills/dexranger
cp SKILL.md ~/.gemini/skills/dexranger/SKILL.md
cp -r scripts ~/.gemini/skills/dexranger/scripts

Verify with /skills or gemini skills list. You may need to restart Gemini CLI to load new skills.

See Gemini CLI Skills docs.


OpenCode

OpenCode discovers skills from multiple locations:

Project-level:

mkdir -p .opencode/skills/dexranger
cp SKILL.md .opencode/skills/dexranger/SKILL.md
cp -r scripts .opencode/skills/dexranger/scripts

User-level:

mkdir -p ~/.config/opencode/skills/dexranger
cp SKILL.md ~/.config/opencode/skills/dexranger/SKILL.md
cp -r scripts ~/.config/opencode/skills/dexranger/scripts

OpenCode also discovers skills from .claude/skills/ and .agents/skills/ directories.

See OpenCode Skills docs.


Cursor

Cursor supports Agent Skills via the .cursor/skills/ directory:

mkdir -p .cursor/skills/dexranger
cp SKILL.md .cursor/skills/dexranger/SKILL.md
cp -r scripts .cursor/skills/dexranger/scripts

Cursor also reads from .claude/skills/ directories.


Other Compatible Agents

This skill works with any agent that supports the Agent Skills open standard. The general pattern is to copy the dexranger/ folder (containing SKILL.md and scripts/) into the agent's skills directory.

Compatible agents include: Amp, Goose, Roo Code, Windsurf, GitHub Copilot, VS Code, Trae, Firebender, Mistral Vibe, Kiro, Factory, Databricks, and many more.

For agents with a .agents/skills/ convention:

mkdir -p .agents/skills/dexranger
cp SKILL.md .agents/skills/dexranger/SKILL.md
cp -r scripts .agents/skills/dexranger/scripts

Standalone (without any AI agent)

Use the script directly from the command line:

python3 scripts/dexranger_check.py <CHAIN> <TOKEN_ADDRESS>

Chains: ETH, SOL, BSC, TON

Output modes:

  • Default: human-readable summary
  • --json: structured JSON (excludes trade activity)
  • --raw: full JSON including trade activity

Examples:

# Check USDT on Ethereum
python3 scripts/dexranger_check.py ETH 0xdac17f958d2ee523a2206206994597c13d831ec7

# Check USDC on Solana (JSON output)
python3 scripts/dexranger_check.py SOL EPjFWdd5AufqSSqeM2qN1xzybapC8G4wEGGkZwyTDt1v --json

Example output:

Token: Tether USD (USDT)
Chain: ETH (Ethereum)
Trust Score: 83/100 (GOOD)
Price: $0.0006162
Market Cap: $579813006
Total Liquidity: $568081.55
DEXes: 5

Warnings (2):
  - Pause/Unpause function present
  - No burn token function

Contract Flags:
  mint: No
  burn: No
  pause: Yes
  blacklist: No
  ...

Requirements

  • curl (pre-installed on macOS, Linux, and most systems)
  • python3 3.8+ (standard library only, no pip packages needed)

Trust Score Reference

Score Rating Meaning
80-100 GOOD Low risk, few or no concerning factors
50-79 WARNING Moderate risk, warning factors found
0-49 DANGER High risk, red flags detected

Links

About

Agent Skill for checking crypto tokens for rug pulls and scams via dexranger.com. Supports ETH, SOL, BSC, TON. Works with Claude Code, Codex, Gemini CLI, Cursor, OpenCode, and 20+ other AI agents.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages