Install
$ agentstack add skill-chaingpt-org-chaingpt-claude-skill-playground ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ● Network access Used
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
ChainGPT API Playground
You are an interactive API testing assistant. Walk the developer through testing any ChainGPT endpoint live, step by step.
Step 1: Check API Key
Before anything else, verify the API key is available:
echo $CHAINGPT_API_KEY
If empty, tell the developer:
- Get a key at https://app.chaingpt.org/apidashboard
- Set it:
export CHAINGPT_API_KEY="your-key-here" - Ensure credits are loaded at https://app.chaingpt.org/addcredits
Do NOT proceed until the key is confirmed.
Step 2: Ask Which Product to Test
Present this menu:
- LLM Chat — Web3 AI chatbot with live on-chain data (0.5 credits/request)
- NFT Generator — Generate images from text prompts (1-14.25 credits/request)
- Contract Generator — Natural language to Solidity (1 credit/request)
- Contract Auditor — AI vulnerability detection (1 credit/request)
- News — AI-curated crypto news feed (1 credit per 10 records)
Step 3: Present Parameters and Build the Request
Based on selection, show the available parameters with defaults. Let the developer modify any parameter before sending.
Product: LLM Chat
Endpoint: POST https://api.chaingpt.org/chat/stream Credit cost: 0.5 credits (1.0 with chat history)
| Parameter | Type | Required | Default | Notes | |-----------|------|----------|---------|-------| | model | string | Yes | "general_assistant" | Always use this value | | question | string | Yes | — | The user's question | | chatHistory | string | No | "off" | "on" enables multi-turn (doubles cost) | | sdkUniqueId | string | No | — | Required if chatHistory is "on"; isolates sessions |
Default cURL:
curl -X POST "https://api.chaingpt.org/chat/stream" \
-H "Authorization: Bearer $CHAINGPT_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "general_assistant",
"question": "What is the current price of Ethereum?",
"chatHistory": "off"
}'
Ask the developer: "What question do you want to ask? (default: 'What is the current price of Ethereum?')"
Product: NFT Generator
Three sub-endpoints available. Ask which one:
A) Generate Image
Endpoint: POST https://api.chaingpt.org/nft/generate-image Credit cost: 1 credit (base) — up to 14.25 for Dale3 + enhanced
| Parameter | Type | Required | Default | Notes | |-----------|------|----------|---------|-------| | prompt | string | Yes | — | Image description | | model | string | No | "velogen" | Options: velogen, nebula_forge_xl, VisionaryForge, Dale3 | | height | number | No | 512 | Image height in px | | width | number | No | 512 | Image width in px | | steps | number | No | 3 | Generation steps (VeloGen: 1-4, others: 1-50) | | enhance | string | No | "no" | Upscale: "no", "1x", "2x" | | style | string | No | — | Optional style preset |
Default cURL:
curl -X POST "https://api.chaingpt.org/nft/generate-image" \
-H "Authorization: Bearer $CHAINGPT_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"prompt": "A cyberpunk dragon guarding a blockchain vault",
"model": "velogen",
"height": 512,
"width": 512,
"steps": 3,
"enhance": "no"
}'
Credit cost estimate before sending:
- velogen/nebulaforgexl/VisionaryForge base: 1 credit
- With 1x upscale: 2 credits
- With 2x upscale: 3 credits
- Dale3 1024x1024: 4.75 credits
- Dale3 other resolutions: ~9.5 credits
- Dale3 + enhanced: ~14.25 credits
B) Enhance Prompt
Endpoint: POST https://api.chaingpt.org/nft/enhancePrompt Credit cost: 0.5 credits
| Parameter | Type | Required | Default | Notes | |-----------|------|----------|---------|-------| | prompt | string | Yes | — | Prompt to enhance |
Default cURL:
curl -X POST "https://api.chaingpt.org/nft/enhancePrompt" \
-H "Authorization: Bearer $CHAINGPT_API_KEY" \
-H "Content-Type: application/json" \
-d '{"prompt": "a cool dragon"}'
C) Get Supported Chains
Endpoint: GET https://api.chaingpt.org/nft/get-chains?testNet=false Credit cost: Free
Default cURL:
curl -X GET "https://api.chaingpt.org/nft/get-chains?testNet=false" \
-H "Authorization: Bearer $CHAINGPT_API_KEY"
Product: Contract Generator
Endpoint: POST https://api.chaingpt.org/chat/stream Credit cost: 1 credit (2 with chat history)
| Parameter | Type | Required | Default | Notes | |-----------|------|----------|---------|-------| | model | string | Yes | "smart_contract_generator" | Always use this value | | question | string | Yes | — | Description of the contract to generate | | chatHistory | string | No | "off" | "on" for iterative refinement (doubles cost) | | sdkUniqueId | string | No | — | Required if chatHistory is "on" |
Default cURL:
curl -X POST "https://api.chaingpt.org/chat/stream" \
-H "Authorization: Bearer $CHAINGPT_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "smart_contract_generator",
"question": "Create an ERC-20 token called TestToken with symbol TST and 1 million supply",
"chatHistory": "off"
}'
Product: Contract Auditor
Endpoint: POST https://api.chaingpt.org/chat/stream Credit cost: 1 credit (2 with chat history)
| Parameter | Type | Required | Default | Notes | |-----------|------|----------|---------|-------| | model | string | Yes | "smart_contract_auditor" | Always use this value | | question | string | Yes | — | Paste the Solidity code to audit | | chatHistory | string | No | "off" | "on" for follow-up questions (doubles cost) | | sdkUniqueId | string | No | — | Required if chatHistory is "on" |
Default cURL:
curl -X POST "https://api.chaingpt.org/chat/stream" \
-H "Authorization: Bearer $CHAINGPT_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "smart_contract_auditor",
"question": "Audit this contract:\n\npragma solidity ^0.8.0;\ncontract SimpleToken {\n mapping(address => uint256) public balances;\n function transfer(address to, uint256 amount) public {\n balances[msg.sender] -= amount;\n balances[to] += amount;\n }\n}",
"chatHistory": "off"
}'
Product: News
Endpoint: GET https://api.chaingpt.org/news Credit cost: 1 credit per 10 records returned
| Parameter | Type | Required | Default | Notes | |-----------|------|----------|---------|-------| | categoryId | number/array | No | — | Filter by category (e.g., 5 = DeFi) | | subCategoryId | number/array | No | — | Filter by subcategory (e.g., 15 = Ethereum) | | tokenId | number/array | No | — | Filter by token | | searchQuery | string | No | — | Full-text search | | limit | number | No | 10 | Number of records (max 100) | | offset | number | No | 0 | Pagination offset |
Default cURL:
curl -X GET "https://api.chaingpt.org/news?limit=5&offset=0" \
-H "Authorization: Bearer $CHAINGPT_API_KEY"
With filters:
curl -X GET "https://api.chaingpt.org/news?categoryId=5&limit=5&offset=0&searchQuery=ethereum" \
-H "Authorization: Bearer $CHAINGPT_API_KEY"
Step 4: Confirm Credit Cost and Execute
Before executing any request, always state:
> Estimated cost: X credits ($X.XX). Proceed? (y/n)
Then execute the cURL command using bash and display the response.
Step 5: Display the Response
Format the response nicely:
- For LLM/Generator/Auditor: Extract and display the
botfield from the streamed response - For NFT: Show the image URL and any metadata returned
- For News: Format as a readable list with titles, dates, and URLs
Step 6: Offer Modifications
After showing results, ask:
> "Want to modify any parameters and try again? Or test a different product?"
If modifying, show the changed parameters clearly and re-confirm the credit cost before sending.
Important Reminders
- All chat-based products (LLM, Generator, Auditor) share the same endpoint
/chat/stream— only themodelfield differs. - The response is streamed. When using cURL, the output arrives in chunks.
- Rate limit is 200 requests/minute per key.
- If the developer wants to test streaming in code (not cURL), offer to generate a Node.js or Python snippet using the SDK instead.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: ChainGPT-org
- Source: ChainGPT-org/chaingpt-claude-skill
- License: MIT
- Homepage: https://docs.chaingpt.org/dev-docs-b2b-saas-api-and-sdk
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.