Skip to content

nexu-io/openmotion / skills/openmotion-media

openmotion-media

Verified

Generate, poll and inspect individual image, video, speech or music assets through openmotion_assets_use after openmotion_media_use discovery (configured Atlas Cloud, fal.ai, WaveSpeedAI or OpenDesign Cloud BYOK providers, or a signed-in OpenDesign account), either when delegated by an OpenMotion film workflow or for a standalone asset-generation request. Native-only tasks with no new speech or music stay outside generation unless explicitly requested. Not a whole-film entry point: creating, recreating, templating, editing or exporting a whole video belongs to the OpenMotion video entries.

OpenMotion Media Generation

Produce verified image, video, speech or music assets that fulfill a specific creative or standalone purpose, returning pinned asset/version identities, provider/model/job identities, measured technical specifications and honest quality observations; import only the selected versions into a film project.

Apply the shared material policy: choose sources by content need, prefer OpenMotion for matching available capabilities, and use external tools for capability gaps. Inside a film task, retain the asset's creative role, use the shared edit workflow for revisions to existing project state, and follow the runtime adapter when continuing project edits.

Apply the shared asset naming contract when assigning names, including standalone media.

Apply the shared generation boundary before discovering or generating media.

Keep credentials strictly isolated: the shared local media service reads saved workspace keys and injects authentication. Never read secrets.json, print or log keys, place credentials into prompts or parameters, or ask the user to paste an API key into chat. Keep user-facing setup, consent and delivery messages in plain language in the user's language; never expose internal Skill names, Tool names, guidance file paths, quoted rule text or internal labels (mg, media, replaced — not scored) in any message to the user.

Provider preference and authorization

Run read-only discovery and obtain consent before submitting any paid generation request:

Choose providers and models yourself; do not ask the user to choose. For each needed asset, use OpenDesign whenever it is available and offers a fitting model: the signed-in OpenDesign account (account) first, then OpenDesign Cloud (opendesign). Only when OpenDesign cannot make that kind of asset, pick among the user's other available providers, preferring the model selected for that kind in defaults when it fits, otherwise the model whose documented inputs and outputs best match the job. A provider or model the user names in this conversation overrides this order. State the chosen provider and model for each asset in the consent question and the delivery reply. Call action: "models", provider, modality, refresh: true (add query to narrow large catalogs) to confirm the choice; catalog visibility establishes discoverability. Live execution establishes account access. Follow setup for the account mediaEntitlement hint: not-included calls for Plus-or-above guidance or another authorized configured provider; unknown proceeds to model discovery only for a signed-in account (available: true). Signed-out and expired sessions omit the hint and require sign-in setup when no other provider is usable. Account available alone does not establish multimodal entitlement.

Ask for spend approval before generating: Unless the user has already approved this exact plan (assets, quantities, provider/model) in the current task, state the planned asset list (modality, count or approximate duration, and selected provider/model) and ask whether they agree to spend their configured multimodal API key or OpenDesign account credits before generating. An approval covers only the assets the user named. When you judge that other shots or assets should also be generated, add them as a separate optional line in the question; if the user already approved generation without further questions, generate only the named assets and offer the others in the delivery reply. Batch this question with any other upfront task questions rather than asking piecemeal per clip. Write it as a plain, direct question in the user's language, one line per planned asset plus one line for any unsupported part and its alternative; stop immediately after the question without explaining why confirmation is needed, and never mention any skill, tool, file path, policy or requirement. Then end the turn and wait for the answer. Do not continue past an unanswered consent question by substituting free, local or code-made voice, music or footage for the planned generated assets, and do not tell the user the question can be ignored. Use such a substitute only when the user chooses it or declines generation.

Read for the current operation

Read the applicable references before their operations; reuse complete bodies already in context. For new generation, read planning, setup and jobs for preparation, including provider schemas, before confirming the plan and authorization above; submit only after authorization, then inspect results. A continuation loads only its affected procedures.

OperationRequired reference
Define or revise an asset for a film or standalone requestPlanning and modality
Discover providers, resolve setup or handle the MCP handbook exampleDiscovery and setup; read-only onboarding needs no project or generation
Submit a new asset, track/resume an existing job, recover a failure or regenerateSubmission and job recovery; new generation uses assets_use, existing media_use jobs retain their original route
Inspect completed candidates, import selected versions or deliver resultsInspection, adoption and handback

MCP-only clients read this entry at openmotion://workflow/media/start; the references are media/planning, media/setup, media/jobs and media/inspection, and each provider guide is media/<provider ID>, listed in the job procedure.

planning-and-task-breakdownaddyosmani/agent-skillsBreaks work into ordered tasks. Use when you have a spec or clear requirements and need to break work into implementable tasks. Use when a task feels too large to start, when you need to estimate scope, or when parallel work is possible.Stars: 119kUpdated Claude CodeCodexVerifiedInstall
screenpipethedotmack/claude-memSearch screen and audio history recorded by Screenpipe and optionally save selected context into claude-mem. Use when the user asks to recall desktop activity, a meeting, or something seen in another app through Screenpipe.Stars: 118kUpdated Claude CodeCodexVerifiedInstall
contact-widgetnexu-io/open-designSelf-contained floating chat widget with welcome screen, social links, meeting button, and message input. Single HTML file, zero dependencies.Stars: 108kUpdated Claude CodeCodexVerifiedInstall
meeting-notesnexu-io/open-designMeeting notes page — title bar with attendees, agenda checklist, decisions block, action items table with owners + dates, and a "next meeting" footer. Use when the brief mentions "meeting notes", "minutes", "1:1 notes", "all-hands recap", or "会议纪要".Stars: 108kUpdated Claude CodeCodexVerifiedInstall