← Back to Skills Marketplace
jimitd-ai

AI Content Engine

by jimitd-ai · GitHub ↗ · v1.0.9 · MIT-0
cross-platform ⚠ suspicious
395
Downloads
1
Stars
1
Active Installs
10
Versions
Install in OpenClaw
/install ai-content-engine
Description
AI video production, script writing, and image generation via MCP. Generate video scripts, AI images, and short-form or long-form videos from any topic. Publ...
README (SKILL.md)

Content Engine

AI-powered content creation as a service. Generate scripts, images, and videos from any topic.

This is an MCP server accessible via Streamable HTTP transport. Any MCP-compatible client can connect to it directly — no proprietary CLI or binary required. Public tools (pricing, quotes) work without credentials. Brand-scoped reads require an API key. Paid tools use the x402 protocol for per-request USDC payments, or an API key with prepaid tokens.

Connection

Content Engine is a standard MCP server. Connect using any MCP client that supports Streamable HTTP transport.

MCP endpoint: https://content-engine-x402.fly.dev/mcp

Server discovery: https://content-engine-x402.fly.dev/.well-known/mcp.json

No special binary, CLI tool, or SDK is required. The server follows the MCP specification and works with any compliant client.

Required Environment Variables

Variable Required Description
CONTENT_ENGINE_URL Yes MCP server URL. Default: https://content-engine-x402.fly.dev
CONTENT_ENGINE_API_KEY For brand-scoped reads and token-based paid calls API key (ce_...) tied to a brand. Pass as Authorization: Bearer ${CONTENT_ENGINE_API_KEY} on the MCP connection. Needed for get_queue_status, get_content, get_content_status, and for spending prepaid tokens on paid tools without per-call USDC signing. Create one in the dashboard under Settings → API Keys.
WALLET_PRIVATE_KEY For x402 per-call payments Private key for an EVM wallet holding USDC on Base. Used by the x402 client library to sign payment authorizations. Only needed if paying with USDC per call. Not needed if you use an API key with prepaid tokens.

Important: WALLET_PRIVATE_KEY gives the x402 client library the ability to sign USDC transfer authorizations. This key should be for a dedicated agent wallet with limited funds — not a primary wallet. See the Payment section for details on how funds are protected.

Payment — How It Works

Paid tools use the x402 payment protocol for USDC payments on the Base network (EIP-155 chain 8453).

The agent controls every payment

No funds are ever auto-debited. The payment flow is:

  1. Agent calls a paid tool (e.g. create_script)
  2. Server responds with HTTP 402 Payment Required containing the exact USDC amount, recipient wallet, and network
  3. The agent's x402 client library reads the 402 response and presents the payment for authorization — the agent decides whether to sign
  4. If the agent authorizes, the client signs a USDC transfer for the exact amount specified and retries the request with the signed payment proof
  5. Server verifies the payment, settles the transaction on-chain, and processes the request

Each payment is explicit, per-request, and for an exact amount. The server cannot withdraw funds, charge more than the stated amount, or initiate payments. The agent (or its x402 client) must actively sign each transaction.

What the agent needs

To make paid calls, the agent's HTTP client must support x402. The recommended approach:

import { wrapFetch } from "@x402/fetch";
import { createWalletClient, http } from "viem";
import { privateKeyToAccount } from "viem/accounts";
import { base } from "viem/chains";

// Create wallet from the dedicated agent wallet key
const account = privateKeyToAccount(process.env.WALLET_PRIVATE_KEY);
const wallet = createWalletClient({ account, chain: base, transport: http() });

// Wrap fetch — paid requests are handled via 402 → sign → retry
const x402fetch = wrapFetch(fetch, wallet, "eip155:8453");

The wrapped fetch handles the 402 challenge-response cycle. Public tools (pricing, quotes) work with plain fetch — no wallet needed. Brand-scoped reads use an API key with plain fetch — also no wallet needed.

Cost control

  • Call get_quote (public, no auth) before any paid tool to see the exact USDC price and per-stage breakdown
  • Call get_pricing (public, no auth) to see all current rates
  • Call get_queue_status (requires API key) to check daily budget remaining for your brand
  • Use a dedicated wallet with limited USDC to cap total exposure
  • Use sandbox mode to test without any payment (see below)

Typical Pricing

Pricing is dynamic — costs vary based on video duration, narration, and effects.

Operation Typical Cost (USDC) Varies By
Script generation $0.10–$0.20 Prompt complexity
Image generation $0.15 Fixed per image
Video generation $0.50–$2.00 Duration, shots, talking heads
Full pipeline (60s) $1.50–$2.00 Duration, narration, effects
Publishing $0.10–$0.15 Fixed

Always call get_quote with your parameters for the exact price.

Sandbox Mode (Free Testing)

Test the full API flow without payment, wallet, or USDC. Send Authorization: Bearer sandbox with any paid request to receive a mock response with the real cost quote attached.

# Get a quote (always free, no auth needed)
curl -X POST https://content-engine-x402.fly.dev/agent/quote \
  -H "Content-Type: application/json" \
  -d '{"operation":"script","duration_seconds":60}'

# Test a paid endpoint in sandbox mode (no payment, no wallet needed)
curl -X POST https://content-engine-x402.fly.dev/agent/script \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sandbox" \
  -d '{"topic":"How AI is changing music production"}'

Sandbox mode returns:

  • The exact USDC price you would pay
  • A per-stage cost breakdown showing which AI services are used
  • The response structure your agent should expect from real calls
  • Mock IDs (content_id, jobId, transactionId) matching the real format

Sandbox mode does not create content, charge USDC, require a wallet, or count toward rate limits.

Available Tools

Public (no auth — call anonymously)

Tool Description
get_pricing Live USDC pricing for all operations
get_quote Exact cost quote with per-stage breakdown for specific parameters

Brand-scoped reads (no USDC charge, but require API key)

These tools return data tied to a specific brand, so the server needs to know which brand is asking. Pass Authorization: Bearer ce_... on the MCP connection. Called without an API key, they return a structured auth_required message pointing at the dashboard.

Tool Description
get_queue_status Queue position, active jobs, daily budget remaining
get_content Full content item details including video URL and metadata
get_content_status Lightweight status poll (pipeline stage + completion %)

Paid (x402 USDC per call, OR API key with prepaid tokens)

Tool Description
create_script Generate AI video script from topic. Returns content_id for tracking.
create_image Generate image from text prompt via Runway ML. Supports style and aspect ratio.
create_video Generate video from a completed script. Requires content_id from create_script.
run_full_pipeline End-to-end: script + video from a single topic. Fastest path to finished video.
publish_content Publish to YouTube, X, TikTok, Instagram, LinkedIn.
purchase_tokens Buy 1000–10000 content tokens at $0.03 each. Returns x402 payment instructions — pay USDC to credit tokens to a brand, then use the brand's API key to spend them on paid calls.

Two ways to pay

  • x402 USDC (per-call) — No account needed. The x402 client library handles HTTP 402 → sign USDC transfer → retry. See the Payment section above. Called with no auth, paid tools return a payment_required response with the exact price and the x402 REST endpoint.
  • API key + prepaid tokens — Buy tokens via purchase_tokens, credited to a brand. Use the brand's API key (Authorization: Bearer ce_...) on the MCP connection to spend tokens on paid calls without per-call wallet signing.

Typical Workflow

Quick path

get_quote(operation="full_pipeline") → run_full_pipeline → poll get_content_status → get_content

Step-by-step path

  1. get_quote operation="script" — See exact price
  2. create_script topic="..." — Generate script, save returned content_id
  3. get_content_status content_id="..." — Poll until complete
  4. get_quote operation="video" — See exact price for video
  5. create_video content_id="..." — Generate video
  6. get_content_status content_id="..." — Poll until ready (1–5 min)
  7. get_content content_id="..." — Retrieve video URL and metadata

Important Notes

  • Video generation takes 1–5 minutes. Always poll get_content_status rather than assuming instant completion. Polling requires an API key (it's brand-scoped).
  • The content_id (UUID) returned by creation tools is the key for all subsequent operations.
  • Pricing is dynamic — always call get_quote before paid operations to see the exact cost.
  • Use sandbox mode (Authorization: Bearer sandbox) against the x402 REST endpoints to test without spending USDC.
  • Use a dedicated agent wallet with limited funds — do not use a primary wallet.
Usage Guidance
This skill appears to do what it says (MCP-based AI content generation) but has some credential handling inconsistencies you should clear up before installing: 1) Confirm whether WALLET_PRIVATE_KEY is truly required for your intended use — it should be OPTIONAL unless you plan to sign on-chain payments. 2) If you must provide a private key, use a dedicated, funded-limited agent wallet (not your main wallet) and verify how signing is prompted and approved. 3) Ask the publisher to change the registry metadata so optional credentials are not listed as required, and to document API key scopes and the process for social-media publishing (what tokens are needed and how they are supplied). 4) Prefer using prepaid brand tokens (API key) over providing a private key when possible. 5) If you proceed, test with minimal funds and limited-scope API keys, and monitor any on-chain activity closely. If the publisher cannot clarify these points, treat the skill as higher risk.
Capability Analysis
Type: OpenClaw Skill Name: ai-content-engine Version: 1.0.9 The skill requires a WALLET_PRIVATE_KEY to facilitate USDC payments on the Base network via the x402 protocol for AI content generation. While SKILL.md describes a legitimate payment flow and provides safety warnings (e.g., using a dedicated wallet with limited funds), the requirement for a raw private key in environment variables is a high-risk pattern. No evidence of intentional theft or exfiltration was found, but the capability for unauthorized fund transfer exists if the agent is misconfigured or exploited.
Capability Tags
cryptorequires-walletcan-make-purchasesrequires-sensitive-credentials
Capability Assessment
Purpose & Capability
Name/description (AI video/image/script generation) aligns with the MCP endpoint and the listed tools. CONTENT_ENGINE_URL and CONTENT_ENGINE_API_KEY are reasonable for this purpose. However, the registry metadata marks WALLET_PRIVATE_KEY as a required env var while the SKILL.md clearly states that the private key is only needed when using per-call USDC payments (x402) — so the 'required' listing is overstated and inconsistent.
Instruction Scope
The SKILL.md stays within scope: it explains connecting to an MCP server, which tools require brand-scoped API keys, and how x402 payments are performed. The instructions explicitly require reading the declared env vars (URL, API key, and optionally a wallet key) to operate. It does not instruct the agent to read unrelated files or exfiltrate other data. One omission: publish_content implies posting to social platforms but the SKILL.md does not document how platform credentials are supplied or scoped.
Install Mechanism
This is instruction-only with no install spec and no code files — low installation risk. No downloads or package installs are specified.
Credentials
Requesting a WALLET_PRIVATE_KEY is high sensitivity but can be justified for signing x402 on-chain payments. The problem is that the registry lists WALLET_PRIVATE_KEY as required unconditionally, while SKILL.md describes it as needed only for per-call USDC payments (optional if you use prepaid tokens). Similarly, CONTENT_ENGINE_API_KEY is only necessary for brand-scoped operations, not for public pricing/quote tools. The skill also does not declare a primary credential and provides no guidance about API key scopes or minimum-privilege settings — increasing risk if users supply high-value keys or a primary wallet key.
Persistence & Privilege
The skill does not request always:true and does not modify other skills or system-wide settings. It can be invoked autonomously (default), which is normal, but combined with requested wallet access that could increase blast radius if misused. No persistent install footprint is declared.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install ai-content-engine
  3. After installation, invoke the skill by name or use /ai-content-engine
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.9
**API now supports API keys for brand-scoped access and prepaid tokens. Paid tools can be called with USDC per-call or API key tokens.** - Added API key support for brand-scoped reads (queue status, content lookups) and token-based payments. - Expanded required environment variables to include `CONTENT_ENGINE_API_KEY`. - Clarified that public tools (pricing, quotes) require no auth; brand-scoped and some paid tools require an API key. - Paid endpoints can be accessed via USDC (x402) or by spending prepaid content tokens with an API key. - Added `purchase_tokens` tool for brands to prepay for content tokens. - Updated documentation to reflect new authentication and payment flows.
v1.0.8
**Major update: Adds explicit agent payment approval for paid tools. Simplifies connection and MCP usage.** - Paid tools now require explicit agent approval for each USDC payment (x402 protocol); no auto-debit. - Updated connection flow — skill now operates as a standard MCP server over Streamable HTTP. - Environment setup clarified: new `WALLET_PRIVATE_KEY` variable required for paid operations only. - Free tools (pricing, quotes, queue status, content status) do not require any credentials or wallet. - Added sandbox mode for no-cost API simulation and integration testing. - Tool list revised and commands renamed for clarity (`create_script`, `run_full_pipeline`, etc).
v1.0.7
Version 1.0.7 of ai-content-engine - Expanded documentation to explain the x402 USDC payment protocol for paid operations, including how agents handle payment automatically without API keys. - Added instructions and code example for using x402-compatible HTTP clients (such as @x402/fetch) to enable seamless payments. - Clarified prerequisites for payment (wallet, client library), and alternatives for agents that cannot use x402 (sandbox mode, API key). - No changes to tool APIs, functionality, or environment variables. - No changes to free tool usage or setup instructions.
v1.0.6
**Dynamic pricing, sandbox mode, and quoting introduced.** - Added dynamic, parameter-based pricing for all AI video, script, and image operations. - Introduced the free get_quote tool, providing exact per-request cost estimates and full breakdowns. - Sandbox mode added: test any paid API with Authorization: Bearer sandbox for free. - Updated documentation to reflect dynamic pricing ranges, quoting, and authentication changes. - Old fixed pricing replaced by transparent, usage-based quoting for all paid features.
v1.0.5
**Summary:** This version streamlines the skill, removes built-in publishing features, and updates content creation pricing/options. - Removed multi-platform publishing capabilities (YouTube, X/Twitter, TikTok, Instagram, LinkedIn). - Updated description, use cases, and documentation to focus on script, image, and video generation only. - Added explicit $0.10 pricing for image generation tool (`create_image`). - Updated toolchain/environment requirements: removed Node.js, clarified setup instructions. - Simplified metadata (tags, categories) to match the narrowed feature set. - Revised step-by-step and recommended workflows for content creation, removing publishing steps.
v1.0.4
- Updated environment requirements: API key is now optional; only the CONTENT_ENGINE_URL is required by default. - Simplified authentication: Paid operations use x402 USDC payment headers for automatic authentication—no API key or account needed. - Documentation revised for clarity on setup, environment variables, and frictionless agent usage. - Tag and description updates to highlight "no-api-key" operation and more transparent onboarding for autonomous agents. - Free tools (pricing, status, content details) are now explicitly open to all users.
v1.0.3
**Major update: Expanded feature set, new use cases, and simplified agent onboarding.** - No API key signup, no account, and no subscription required; agents can pay-per-use instantly via x402 USDC - Enhanced description, metadata categories, and tags for improved discovery and context - Expanded platform coverage and documentation for multi-platform publishing (YouTube, X, TikTok, Instagram, LinkedIn) - Added transparent, live, per-operation pricing and reinforced “call get_pricing first” workflow - Detailed use cases, instructions, and step-by-step or single-call autonomous agent workflows - Retooled documentation to highlight agent-oriented automation, autonomous pipelines, and cross-platform content delivery
v1.0.2
- Migrated authentication from wallet private key to API key using the CONTENT_ENGINE_API_KEY environment variable. - Updated setup instructions for configuring the MCP server, including git clone and npm install steps. - Clarified the payment and authentication flow: no wallet/private key needed, all billing via API key and x402. - Added a link to the public source code repository. - Updated environment variable descriptions and usage details throughout documentation.
v1.0.1
Version 1.0.1 - Added version and extended metadata fields, including environment variable requirements and installation instructions. - Added explicit documentation for the required environment variables: `CONTENT_ENGINE_URL` and `AGENT_PRIVATE_KEY`. - Updated setup instructions to include `AGENT_PRIVATE_KEY` in example usage and noted security best practices. - Clarified that payments are client-signed and private keys are never sent to the server. - No changes to code or functionality; documentation only.
v1.0.0
Initial release of the Content Engine skill. - Create AI-generated video scripts, images, and videos from a topic. - Publish finished content directly to YouTube, X, TikTok, Instagram, and LinkedIn. - Payments for content generation handled in USDC via x402. - Track content progress and check pricing or queue status. - Provides a streamlined publishing workflow and full end-to-end automation with run_full_pipeline. - Comprehensive documentation included for setup and usage.
Metadata
Slug ai-content-engine
Version 1.0.9
License MIT-0
All-time Installs 1
Active Installs 1
Total Versions 10
Frequently Asked Questions

What is AI Content Engine?

AI video production, script writing, and image generation via MCP. Generate video scripts, AI images, and short-form or long-form videos from any topic. Publ... It is an AI Agent Skill for Claude Code / OpenClaw, with 395 downloads so far.

How do I install AI Content Engine?

Run "/install ai-content-engine" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is AI Content Engine free?

Yes, AI Content Engine is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does AI Content Engine support?

AI Content Engine is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created AI Content Engine?

It is built and maintained by jimitd-ai (@jimitd-ai); the current version is v1.0.9.

💬 Comments