video-generator

Executes approved video-generation requests through provider APIs with polling, manifests, and cost tracking.

Updated Jun 12, 2026
One-click install
npx skills add https://github.com/oleyna80/agentic-sdlc-framework --skill video-generator-oleyna80
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: video-generator
Source: https://github.com/oleyna80/agentic-sdlc-framework/tree/main/skills/video-generator
Command: npx skills add https://github.com/oleyna80/agentic-sdlc-framework --skill video-generator-oleyna80

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Running AI video generation through provider APIs involves fragile async job submission, polling, credential handling, and cost control. This Skill executes only pre-approved generation packages through a governed adapter contract, keeping every output traceable and within budget. ## Core Features & Use Cases - Provider adapter contract: Standardized validate-config, estimate, submit, status, download, and cancel operations with normalized JSON output. - Governed execution: Enforces preconditions (approved brief, rights report, prompt package, budget) and stops if credentials leak into tracked files. - Traceable manifests: Records request IDs, parameters, costs, and terms verification for every generated asset. - Use Case: A content team has an approved prompt package and budget for three candidate clips. The Skill validates credentials, estimates cost against the limit, submits exactly three jobs, polls with bounded backoff, downloads and verifies the outputs, and records actual spend. ## Quick Start Run the approved video generation package in the workspace and produce the candidates with a cost manifest.

Frequently Asked Questions about video-generator

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run an AI video generation job through a provider API?▼

Approve the brief, rights report, prompt package, provider, and budget first, then invoke the Skill with the approved package. It validates credentials, estimates cost, submits the exact candidate count, polls with bounded backoff, and downloads verified outputs.

What must a video provider adapter support?▼

The adapter must implement validate-config, estimate, submit, status, download, and cancel operations. It accepts a provider-neutral request and writes normalized JSON, with provider-specific parameters kept in the request manifest.

Can I retry a failed video generation request?▼

Transient transport failures may be retried at most twice, using idempotency keys when supported. Requests with an existing ID are not resubmitted until resolved, and moderation or rights failures are never retried through another provider.

How are API keys and credentials handled during generation?▼

Credentials are read from environment variables or an approved secret manager and never printed. Authorization headers, signed URLs, tokens, and account IDs are redacted from logs, and the Skill stops if a credential appears in a tracked file.

What happens if generation exceeds the approved budget?▼

A cost estimate runs before submission and is compared against the approved limit. The Skill never silently increases candidate count, duration, resolution, or cost, and actual spend is recorded in the manifest.