# GenCreator Skill Pack Library

Canonical: https://gencreator.ai/products/skill-packs
API: https://gencreator.ai/api/products/skill-packs

A portable library of creator operating skills that installs GenCreator memory, signal, creation, approval, community, and Eve orchestration into assistant workbenches.

First win: Choose one skill pack, install one mode-safe workflow, and produce one usable artifact or approval record in the first sitting.

Support boundary: The library includes install files, prompts, lessons, MCP resources, and support boundaries. It does not include autonomous publishing, private implementation, paid entitlement access, or automatic community posting.

## Install Order

1. Creator Vault Librarian: Give the assistant memory and proof before output.
2. Outlier Signal Analyst: Decode what is working without copying surface style.
3. Campaign Pack Producer: Draft one campaign packet from an approved signal.
4. Approval Flow Operator: Create the decision record before any public action.
5. Community Proof Concierge: Move finished artifacts into proof-based practice.
6. Eve Orchestrator Seed: Coordinate the loop only after the pieces are understandable.

## Skill Packs

## Creator Vault Librarian

ID: creator-vault-librarian
Stage: memory
Mode: read_only
Surface: Creator Vault
Open: https://gencreator.ai/products/delivery#creator-vault-kit
Buy or inspect: https://gencreator.ai/products/buy?product=creator-vault-kit
Checkout contract: https://gencreator.ai/products/checkout?product=creator-vault-kit
Install time: 8 minutes

Keep voice, offer, audience, proof, and approval policy clean before the assistant writes.

First win: Fill one memory row and return the missing proof questions before a campaign is drafted.

Products:
- Creator Vault Kit: https://gencreator.ai/products/delivery#creator-vault-kit
- Agent and MCP Pack: https://gencreator.ai/products/agent-pack

MCP resources:
- gencreator-product-manifest: https://gencreator.ai/api/products/manifest
  Use: Confirm product status and support boundary.
- creator-vault-kit-starter: https://gencreator.ai/downloads/gencreator/creator-vault-kit-starter.md
  Use: Read the starter schema before creating memory fields.

Prompts:
- Create the first vault row: Read the artifact summary. Return voice rule, audience, offer path, proof metric, approval policy, and missing proof questions. Do not draft public copy.
  Output: One complete vault row plus missing-proof questions.

Files:
- gencreator-creator-vault-librarian.skill.md: Install the role instructions and run the starter prompt.
- gencreator-creator-vault-librarian.lesson.md: Teach the creator the first workflow before adding automation.
- gencreator-creator-vault-librarian.resources.json: Wire the assistant to read the right GenCreator resources before acting.

Lesson:
- Title: Memory before output
- Objective: Teach the creator why assistants need voice, offer, proof, and policy before drafting.
- Exercise: Extract one real sentence from the creator and turn it into a voice rule.
- Proof: One filled vault row with a source note.

Community practice: Share the memory row and ask whether the proof field is strong enough for public claims.

Boundaries:
- Do not invent proof or testimonials.
- Do not paste secrets or private customer data into the vault row.
- Do not draft public content until missing proof is named.

Support boundary: Self-serve memory setup. Custom vault migration belongs in Studio.


## Outlier Signal Analyst

ID: outlier-signal-analyst
Stage: signal
Mode: read_only
Surface: Outlier Radar
Open: https://gencreator.ai/radar
Buy or inspect: https://gencreator.ai/products/buy?product=campaign-pack-system
Checkout contract: https://gencreator.ai/products/checkout?product=campaign-pack-system
Install time: 10 minutes

Turn public performance signals into hook patterns and creator-specific recommendations.

First win: Decode one outlier and state the mechanism before creating any new draft.

Products:
- Campaign Pack System: https://gencreator.ai/products/delivery#campaign-pack-system
- Agent and MCP Pack: https://gencreator.ai/products/agent-pack

MCP resources:
- gencreator-creation-blueprint: https://gencreator.ai/api/blueprint
  Use: Read the creator type, output path, and first artifact.
- gencreator-product-manifest: https://gencreator.ai/api/products/manifest
  Use: Confirm which product should receive the signal.

Prompts:
- Decode one outlier signal: Given one public post, video, or campaign result, identify the hook pattern, mechanism, proof used, audience tension, and safest next experiment.
  Output: One signal brief with mechanism, proof note, and next experiment.

Files:
- gencreator-outlier-signal-analyst.skill.md: Install the role instructions and run the starter prompt.
- gencreator-outlier-signal-analyst.lesson.md: Teach the creator the first workflow before adding automation.
- gencreator-outlier-signal-analyst.resources.json: Wire the assistant to read the right GenCreator resources before acting.

Lesson:
- Title: Signals are not scripts
- Objective: Teach the creator to extract mechanisms from outliers without copying surface style.
- Exercise: Compare one high-performing post to one baseline post and name the difference.
- Proof: A signal brief with one mechanism and one constraint.

Community practice: Bring the decoded mechanism and ask which proof would make it usable for the creator.

Boundaries:
- Do not scrape private communities.
- Do not copy another creator’s wording as final output.
- Do not treat view count as proof of offer fit.

Support boundary: Self-serve signal analysis. Custom analytics ingestion is a later product milestone.


## Campaign Pack Producer

ID: campaign-pack-producer
Stage: creation
Mode: draft_only
Surface: Campaign Studio
Open: https://gencreator.ai/campaigns
Buy or inspect: https://gencreator.ai/products/buy?product=campaign-pack-system
Checkout contract: https://gencreator.ai/products/checkout?product=campaign-pack-system
Install time: 12 minutes

Turn one approved signal into platform drafts, scripts, replies, and an approval question.

First win: Create one campaign packet with the same mechanism across LinkedIn, X, newsletter, and short script.

Products:
- Campaign Pack System: https://gencreator.ai/products/delivery#campaign-pack-system
- Agent and MCP Pack: https://gencreator.ai/products/agent-pack

MCP resources:
- gencreator-self-serve-kit: https://gencreator.ai/api/products/kit
  Use: Read the generated product handoff and support boundary.
- campaign-pack-system-starter: https://gencreator.ai/downloads/gencreator/campaign-pack-system-starter.md
  Use: Use the starter packet as the campaign structure.

Prompts:
- Generate an approval-safe campaign pack: Use one signal, one proof metric, and one offer path. Draft platform variants, label missing proof, and end with one human approval question.
  Output: A multi-platform campaign packet with draft-only risk notes.

Files:
- gencreator-campaign-pack-producer.skill.md: Install the role instructions and run the starter prompt.
- gencreator-campaign-pack-producer.lesson.md: Teach the creator the first workflow before adding automation.
- gencreator-campaign-pack-producer.resources.json: Wire the assistant to read the right GenCreator resources before acting.

Lesson:
- Title: One mechanism, many formats
- Objective: Teach the creator to preserve the same idea across platforms instead of making disconnected posts.
- Exercise: Draft four formats from one mechanism and highlight the proof used in each.
- Proof: A campaign packet with one approval question.

Community practice: Share the campaign packet and ask which claim needs stronger proof before publishing.

Boundaries:
- Do not publish or schedule automatically.
- Do not make product claims beyond manifest status.
- Do not remove the approval question.

Support boundary: Self-serve drafting. Publishing, scheduling, and customer messaging remain approval-required.


## Approval Flow Operator

ID: approval-flow-operator
Stage: approval
Mode: approval_required
Surface: Approval Queue
Open: https://gencreator.ai/approvals
Buy or inspect: https://gencreator.ai/products/buy?product=agent-mcp-pack
Checkout contract: https://gencreator.ai/products/checkout?product=agent-mcp-pack
Install time: 7 minutes

Convert public, customer-facing, paid, or reputation-sensitive actions into human decision records.

First win: Create one approval record with action, sources, risk, human decision, and next handoff.

Products:
- Agent and MCP Pack: https://gencreator.ai/products/agent-pack

MCP resources:
- gencreator-agent-pack-bundle: https://gencreator.ai/api/products/agent-pack/bundle
  Use: Read the approval policy and role files.
- gencreator-checkout-readiness: https://gencreator.ai/api/products/checkout
  Use: Confirm checkout and fulfillment status before sales claims.

Prompts:
- Prepare an approval record: For the proposed action, list source artifacts, risk level, human decision needed, support-boundary note, and safe next action. Do not execute the action.
  Output: A decision record ready for human approval.

Files:
- gencreator-approval-flow-operator.skill.md: Install the role instructions and run the starter prompt.
- gencreator-approval-flow-operator.lesson.md: Teach the creator the first workflow before adding automation.
- gencreator-approval-flow-operator.resources.json: Wire the assistant to read the right GenCreator resources before acting.

Lesson:
- Title: Trust is a workflow
- Objective: Teach the creator when an assistant must stop and ask for a decision.
- Exercise: Convert one risky action into an approval record.
- Proof: An approval record with clear risk and next action.

Community practice: Share the approval record and ask whether the support boundary is clear enough.

Boundaries:
- Do not execute public or paid actions automatically.
- Do not approve on behalf of the creator.
- Do not hide risk notes to make a draft feel finished.

Support boundary: Self-serve approval patterns. Legal, finance, and regulated claims need qualified review.


## Community Proof Concierge

ID: community-proof-concierge
Stage: community
Mode: draft_only
Surface: GenCreator.community
Open: https://gencreator.ai/community/onboarding
Buy or inspect: https://gencreator.ai/products/buy?product=gencreator-community
Checkout contract: https://gencreator.ai/products/checkout?product=gencreator-community
Install time: 9 minutes

Turn finished artifacts into precise community review requests and weekly proof loop updates.

First win: Draft one review request with artifact, proof metric, what changed, and one decision needed.

Products:
- GenCreator.community: https://gencreator.ai/community/os
- Agent and MCP Pack: https://gencreator.ai/products/agent-pack

MCP resources:
- gencreator-creation-workspace-session: https://gencreator.ai/api/workspace/session
  Use: Read the artifact, proof metric, checkout handoff, and community packet.
- gencreator-self-serve-kit: https://gencreator.ai/api/products/kit
  Use: Read the generated modules and activation rhythm.

Prompts:
- Draft a proof-loop review request: Create a review request with artifact summary, proof available, what changed, one decision needed, and the next action after review.
  Output: A community-ready review request that asks for one decision.

Files:
- gencreator-community-proof-concierge.skill.md: Install the role instructions and run the starter prompt.
- gencreator-community-proof-concierge.lesson.md: Teach the creator the first workflow before adding automation.
- gencreator-community-proof-concierge.resources.json: Wire the assistant to read the right GenCreator resources before acting.

Lesson:
- Title: Practice after artifact
- Objective: Teach the creator to bring finished work into review instead of asking broad questions.
- Exercise: Rewrite broad feedback into one decision question.
- Proof: A review request tied to one artifact.

Community practice: Post the review request after membership and access are confirmed.

Boundaries:
- Do not post into a community automatically.
- Do not ask for broad feedback when one decision is needed.
- Do not claim membership access is active unless the manifest or account state proves it.

Support boundary: Self-serve community prep. Automatic posting waits for gated member workflows.


## Eve Orchestrator Seed

ID: eve-orchestrator-seed
Stage: orchestration
Mode: approval_required
Surface: Eve Workbench
Open: https://gencreator.ai/eve
Buy or inspect: https://gencreator.ai/products/buy?product=agent-mcp-pack
Checkout contract: https://gencreator.ai/products/checkout?product=agent-mcp-pack
Install time: 15 minutes

Coordinate memory, signal, campaign, approval, and community handoff as one operating loop.

First win: Choose the right next skill pack, run it in the safest mode, and prepare the human approval task.

Products:
- Agent and MCP Pack: https://gencreator.ai/products/agent-pack
- Self-Serve Kit Composer: https://gencreator.ai/products/kit
- GenCreator.community: https://gencreator.ai/community/os

MCP resources:
- gencreator-eve-operating-brief: https://gencreator.ai/api/eve/brief
  Use: Read role stack, MCP plan, approval task, and community handoff.
- gencreator-skill-pack-library: https://gencreator.ai/api/products/skill-packs
  Use: Choose the next skill pack and install files.
- gencreator-product-manifest: https://gencreator.ai/api/products/manifest
  Use: Confirm product status before routing the buyer.

Prompts:
- Select the next skill pack: Read the artifact state, product manifest, and skill library. Choose the next skill pack, explain the mode, create the first task, and name the approval boundary.
  Output: A routed operating task with mode, skill pack, approval note, and handoff.

Files:
- gencreator-eve-orchestrator-seed.skill.md: Install the role instructions and run the starter prompt.
- gencreator-eve-orchestrator-seed.lesson.md: Teach the creator the first workflow before adding automation.
- gencreator-eve-orchestrator-seed.resources.json: Wire the assistant to read the right GenCreator resources before acting.

Lesson:
- Title: Eve as operating loop
- Objective: Teach the creator to treat Eve as orchestration, not a mascot or autonomous publisher.
- Exercise: Route one artifact through the safest next skill pack.
- Proof: A mode-safe Eve task with one approval boundary.

Community practice: Share the routed task and ask whether the next skill pack matches the artifact stage.

Boundaries:
- Do not impersonate autonomous judgment.
- Do not take actions outside the chosen mode.
- Do not claim the full superagent is live before persistence and entitlement gates exist.

Support boundary: Eve remains an operating role until persistence, entitlement, and community handoff are wired.


## Education Track

- Memory before output (Creator Vault Librarian): One filled vault row with a source note.
- Signals are not scripts (Outlier Signal Analyst): A signal brief with one mechanism and one constraint.
- One mechanism, many formats (Campaign Pack Producer): A campaign packet with one approval question.
- Trust is a workflow (Approval Flow Operator): An approval record with clear risk and next action.
- Practice after artifact (Community Proof Concierge): A review request tied to one artifact.
- Eve as operating loop (Eve Orchestrator Seed): A mode-safe Eve task with one approval boundary.

## Community Practice

- Creator Vault Librarian: Share the memory row and ask whether the proof field is strong enough for public claims.
- Outlier Signal Analyst: Bring the decoded mechanism and ask which proof would make it usable for the creator.
- Campaign Pack Producer: Share the campaign packet and ask which claim needs stronger proof before publishing.
- Approval Flow Operator: Share the approval record and ask whether the support boundary is clear enough.
- Community Proof Concierge: Post the review request after membership and access are confirmed.
- Eve Orchestrator Seed: Share the routed task and ask whether the next skill pack matches the artifact stage.

## MCP Resources

- gencreator-skill-pack-library: https://gencreator.ai/api/products/skill-packs
  Use: Read installable skill packs, lessons, prompts, and boundaries.
- gencreator-agent-pack-bundle: https://gencreator.ai/api/products/agent-pack/bundle
  Use: Install the broader role files, MCP config, approval policy, and Eve operator.
- gencreator-self-serve-kit: https://gencreator.ai/api/products/kit
  Use: Generate a buyer-ready product packet that chooses the next skill pack.
- gencreator-eve-operating-brief: https://gencreator.ai/api/eve/brief
  Use: Route the artifact into a mode-safe Eve task.