Update to Creator Workspace

{
  "description": "Route requests across Creator Workspace to the smallest useful set of bundled skills. Use whenever the user explicitly invokes Creator Workspace, asks for a multi-step creator workflow, or the right second-brain, recall, writing, social-media, visualization, analytics, or session-learning skill is unclear.",
  "included_files": [
    {
      "relative_path": "agents/openai.yaml",
      "size_in_bytes": 295
    },
    {
      "relative_path": "references/routing-graph.md",
      "size_in_bytes": 2515
    },
    {
      "relative_path": "references/workspace-contract.md",
      "size_in_bytes": 841
    }
  ],
  "name": "creator-router",
  "skill_md_contents": "---\nname: creator-router\ndescription: Route requests across Creator Workspace to the smallest useful set of bundled skills. Use whenever the user explicitly invokes Creator Workspace, asks for a multi-step creator workflow, or the right second-brain, recall, writing, social-media, visualization, analytics, or session-learning skill is unclear.\n---\n\n# Creator Router\n\nTreat an explicit `@Creator Workspace` mention as a direct invocation of this router. Continue into the most specific bundled skill instead of merely describing what the plugin could do.\n\n## Principle\n\nRoute by the user's finished outcome, not by upstream repository names. Use the fewest skills needed, but preserve required handoffs between memory, voice, creation, visuals, analytics, and explicit persistence.\n\n## Primary routes\n\nKnowledge and context:\n- create or adapt a second brain -> `second-brain-setup`\n- import AI conversation exports -> `conversation-importer`\n- retrieve existing workspace knowledge or resume prior work -> `workspace-recall`\n- synthesize or repair linked knowledge -> `living-wiki`\n- daily context or idea brief -> `brain-briefs`\n- create content grounded in stored context -> `create-from-brain`\n\nVoice and authorship:\n- learn style from written samples -> `voice-builder`\n- preserve literal spoken authorship -> `voiceprint`\n- newsletter-specific style -> `newsletter-voice`\n\nSocial creation:\n- draft -> `post-writer`\n- structure -> `post-formatter`\n- hooks -> `hook-generator`\n- idea bank -> `content-matrix`\n- current topic research -> `niche-research`\n- LinkedIn profile -> `profile-optimizer`\n- graphic -> `graphic-designer`\n- infographic -> `infographic-generator`\n- carousel -> `carousel-generator`\n- quote post -> `quote-post`\n- pinned comment -> `pinned-comment`\n- short-form video -> `reels-scripting`\n- YouTube thumbnail -> `youtube-thumbnail`\n- draft scoring -> `post-scorer`\n- LinkedIn analytics -> `analytics-dashboard`\n\nVisual handoff and persistence:\n- visual plans, comparisons, reports, or option boards -> `show-me`\n- save explicit corrections, durable preferences, project state, or lessons -> `save-progress`\n\nExecution support:\n- workspace discovery, reads, writes, patches, shell, or repository state -> `host-workspace-operator`\n- deterministic parsing, indexing, provenance, analytics, hashing, or validation -> `sandbox-python-executor`\n\n## Neural handoff rules\n\n1. If a request depends on stored context, run `workspace-recall` before creation unless the required source is already supplied in the current turn.\n2. If imported material is raw, use `conversation-importer` or `living-wiki` before treating it as synthesized knowledge.\n3. If voice fidelity matters, apply `voice-builder` after recall and before drafting. Use `voiceprint` instead when the user wants literal spoken wording preserved.\n4. Route creation to the narrowest production skill after context and voice are resolved.\n5. Add `show-me` only when a visual decision surface or artifact is useful.\n6. Use `sandbox-python-executor` when correctness depends on real computation rather than prose reasoning.\n7. Use `host-workspace-operator` for actual workspace I/O and never claim an operation occurred without host evidence.\n8. Never invoke `save-progress` merely because work finished. Persist only when the user explicitly asks to retain learning or project state.\n\n## Composite examples\n\n\"Use my notes to write a LinkedIn post in my voice\":\n`workspace-recall -> voice-builder profile -> post-writer`\n\n\"Turn these exports into a second brain and tell me what you know about topic X\":\n`second-brain-setup -> conversation-importer -> living-wiki -> workspace-recall`\n\n\"Analyze my LinkedIn data and show me what is working\":\n`analytics-dashboard -> sandbox-python-executor -> show-me`\n\n\"Create a post from my research, then save the correction I made to my tone\":\n`workspace-recall -> create-from-brain -> post-writer -> save-progress`\n\n## Context precedence\n\nPrefer, in order when relevant:\n1. current-turn user material\n2. current project state\n3. applicable durable rules and voice profile\n4. synthesized wiki knowledge\n5. raw/imported source material for provenance or missing details\n\nDo not silently turn one-off instructions into global rules.\n\n## Evidence rule\n\nNever say a file was read, the web was searched, Python ran, an image rendered, a browser opened, memory changed, or a connector was queried unless the current host produced evidence.\n\n## References\n\nRead `references/workspace-contract.md` for storage ownership and `references/routing-graph.md` for cross-skill handoffs.\n"
}