More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    AI Video Generation Package Compiler

    1

    Convert approved shot contracts into validated, capability-aware prompt packages for Dreamina, Flow, or Runway.

    $11

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    AI Video Generation Package Compiler

    AI Video Generation Package Compiler

    Example session with this skill installed

    Compile the approved SH-002 shot contract for Google Flow and Runway using the supplied dated capability snapshot, approved reference registry, and routing requirements: 9:16, six seconds, image-to-video, silent generation with post-produced audio.

    • Read your context and instructions
    • Compiled the ai video generation

    Demonstration Output — Generation Package

    Execution state: prepared_not_executed
    Authorization state: not_requested
    No provider was called and no credits were spent.

    Routing decision
    SH-002 is ROUTABLE to the supplied Google Flow and Runway fixture capabilities. Dreamina was not requested and is not silently added. The compiler applied hard filters in order: interface availability → mode → reference roles → duration → aspect ratio → resolution → audio intent → rights and approval.

    Reference assignments
    REF-CHAR-001 → identity role; approved, rights-approved, in scope for SH-002.
    REF-WORLD-001 → environment role; approved, rights-approved, in scope for SC-002.
    No unapproved reference was restored after filtering.

    Canonical shot lock
    Primary action: drag one quoted comment into its matching evidence row.
    Start state: three unclassified comments.
    End state: one comment linked to source, scope, and owner.
    Continuity: navy overshirt, silver glasses, left-wrist watch, fixed desk axis, daylight from frame left.

    Google Flow manifest row
    Mode: image-to-video · Duration: 6s · Aspect: 9:16 · Audio: post. Prompt fragments preserve subject, action, environment, camera, lighting, continuity locks, and exclusions as separate mapped fields.

    Runway manifest row
    Mode: image-to-video · Duration: 6s · Aspect: 9:16 · Audio: post. Platform syntax changes, but the canonical action and end state remain identical.

    Validation receipt
    PASS — each approved shot appears exactly once per requested platform; every manifest row resolves to an approved reference and current supplied capability record; prompt-pack hash matches; no execution or authorization claim appears. This receipt proves package structure and traceability only, not provider behavior or visual quality.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Copying one prompt across AI-video platforms hides critical differences in reference roles, duration, aspect ratio, resolution, audio support, and interface constraints. AI Video Generation Package Compiler turns approved shot contracts into dated, platform-specific Generation Packages for Dreamina, Google Flow, or Runway. It filters candidates against supplied capability evidence, assigns only approved reference assets, preserves every semantic fragment from the canonical shot, and produces explicit UNROUTABLE records instead of inventing compatibility. What it delivers: - One execution manifest per requested platform - Platform-dialect prompts mapped back to canonical shot fields - Reference-role assignments with approval and rights checks - Capability and freshness gates for mode, duration, ratio, resolution, and audio - A prompt pack plus deterministic validation receipt Boundaries: it does not browse provider documentation, authenticate, call providers, generate media, spend credits, edit outputs, upload, or publish. Packages remain prepared_not_executed until a separate authorized operator executes them.

    Requires local capability-snapshot.json and reference-registry.json files. Does not handle credit spending, media uploads, or live provider authentication.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 2 months ago

    Permissions

    Terminal / Shell
    Listed2 months ago

    What's inside

    Frequently Asked Questions

    This skill prevents generation failures by translating generic shot ideas into platform-specific 'Generation Packages' that respect the unique technical constraints (ratio, duration, and asset roles) of Dreamina, Flow, and Runway.

    The skill supports Dreamina, Google Flow, and Runway Gen-2/Gen-3; it uses capability-aware logic to ensure your shots are only assigned to platforms that can actually support the requested resolution and audio specs.

    No, this skill is a preparation and validation layer; it generates a deterministic 'prompt pack' and execution manifest that you or another agent then use to trigger the actual generation on your preferred platform.

    You receive the compiler skill itself, which produces platform-dialect prompts, reference asset assignments, and a validation receipt that confirms the shot meets current provider capabilities.

    If a requested shot violates a platform's current capabilities, the skill produces an 'UNROUTABLE' record, preventing wasted credits on impossible generations rather than guessing or inventing compatibility.

    To start, you need your approved 'shot contracts' or canonical shot fields and any reference assets you plan to use; the skill will then filter these against its internal capability gates to build your packages.

    New in Image & Video