feat(capability): define Discord image generation and asset storage flow

[egdev6]

Pre-flight checks

• I searched existing issues and this is not a duplicate
• I understand this issue must be approved before implementation starts
• I will not include real Discord IDs, tokens, credentials, screenshots, raw logs, transcripts, or private payloads in public evidence

Problem

The user wants image creation from Discord with filesystem access. OpenClaw has several media/model provider plugins and ClawHub image candidates, but the project needs a workflow for prompts, approvals, output storage, provenance, and private filesystem boundaries.

Proposed solution

Define a Discord image generation capability and workflow. Select candidate providers/plugins, define where generated assets are stored, require scoped filesystem permission for saving outside the workspace, and pair with a skill for brief → draft → review → approved asset.

Area

Runtime

Acceptance criteria

• Candidate image generation providers/plugins are documented and verified in the current runtime or via ClawHub search.
• Asset storage target is classified as repo-backed, private runtime, or local filesystem.
• Filesystem writes require explicit scope permission and approval.
• Generated asset metadata includes prompt summary, provider, timestamp, and intended use.
• Fake-first flow avoids real private images and account data.
• Discord response previews are bounded and safe.

Validation plan

• Validate provider/plugin availability without generating private assets.
• Use demo prompts and fake metadata fixtures.
• Confirm no file write happens before approval.

Additional context

Bundled providers include fal, comfy, openai, and others; ClawHub image candidates exist but need vetting.