# GenCreator Activation Plan Activation: activation_96f25f99 Canonical: https://gencreator.ai/activate Founder-led creator activation for agent install: make the first artifact useful, then turn it into a governed weekly system before selling higher-touch support. ## First Win Seven-day sprint path: finish point-of-view post, create one proof check, install one role or kit module, and route one precise review question. ## Transformation Frame GenCreator activation is the bridge from useful artifact to repeatable operating behavior: create the packet, package the kit, install the agent habit, ask for proof, and enter community only with something real to review. ## Value Stack - Useful artifact: Founder-led creator leaves with a creation packet instead of another abstract plan. Proof: Creation Engine packet, draft artifact, and proof check. - Scalable product: The packet becomes kit modules, assistant prompts, MCP resources, and checkout handoff. Proof: Self-Serve Kit contract and product manifest status. - Agent habit: Roles and tools are mode-safe before any public or paid action. Proof: Skill Pack Library, Agent Pack bundle, and approval policy. - Community practice: GenCreator.community starts from a finished artifact and one decision. Proof: Artifact-first onboarding and weekly proof loop. ## Product Path 1. Creation Engine - Open: https://gencreator.ai/create - Buy or inspect: https://gencreator.ai/create - Buyer action: Free self-serve tool: open now and export the artifact. - Commercial stage: free product surface - Provider status: not_required - Why: The first product action is creating proof, not diagnosing forever. 2. Self-Serve Kit Composer - Open: https://gencreator.ai/products/kit - Buy or inspect: https://gencreator.ai/products/kit - Buyer action: Free self-serve tool: open now and export the artifact. - Commercial stage: free product surface - Provider status: not_required - Why: The packet needs modules, prompts, resources, and support boundaries. 3. Skill Pack Library - Open: https://gencreator.ai/products/skill-packs - Buy or inspect: https://gencreator.ai/products/skill-packs - Buyer action: Free self-serve tool: open now and export the artifact. - Commercial stage: free product surface - Provider status: not_required - Why: One skill gives the assistant a governed workflow. 4. Agent and MCP Pack - Open: https://gencreator.ai/products/agent-pack - Buy or inspect: https://gencreator.ai/products/buy?product=agent-mcp-pack - Buyer action: Use the preview route now; paid checkout opens after provider product wiring. - Commercial stage: paid digital packet pending checkout wiring - Provider status: needs_provider_product - Why: Roles, MCP config, Eve, and approval policy turn the work into safer software. 5. GenCreator.community - Open: https://gencreator.ai/community/os - Buy or inspect: https://gencreator.ai/products/buy?product=gencreator-community - Buyer action: Preview the proof-loop entry; paid Circle access waits for Whop and Skool gating. - Commercial stage: membership preview before gated launch - Provider status: planned - Why: The community should review a real artifact and one decision. 6. Studio - Open: https://gencreator.ai/studio/apply - Buy or inspect: https://gencreator.ai/studio/apply - Buyer action: Apply only after the activation plan has proof and a specific review need. - Commercial stage: capped Studio application - Provider status: application_ready - Why: Direct review belongs after the self-serve system is already producing proof. ## Seven-Day Activation ### Day 1: Create the packet Job: Turn the source signal into a first artifact before choosing more support. Route: https://gencreator.ai/create Product: Creation Engine Output: Point-of-view post Proof: Method, constraint, and result from the source signal. MCP resource: gencreator-creation-engine Support boundary: Self-serve creation before community or Studio. ### Day 2: Package the kit Job: Translate the packet into modules, prompts, resources, and checkout-aware handoff. Route: https://gencreator.ai/products/kit Product: Self-Serve Kit Composer Output: Agent-first GenCreator kit for Founder-led AI creation system Proof: Install one assistant role, read one MCP resource, and create eve_e2181b65 before any public output. MCP resource: gencreator-self-serve-kit Support boundary: The kit includes setup prompts, starter artifacts, API contracts, and update paths. It does not include private implementation, paid entitlement access, or automated community posting until those systems are wired. ### Day 3: Install one role Job: Give the assistant a single mode-safe job before expanding automation. Route: https://gencreator.ai/products/skill-packs Product: Skill Pack Library Output: Agent install task Proof: Role mode is read_only, draft_only, or approval_required. MCP resource: gencreator-agent-pack-bundle Support boundary: Agents inspect and draft; humans approve external actions. ### Day 4: Prepare approval Job: Convert the riskiest public, paid, or customer-facing move into a review record. Route: https://gencreator.ai/approvals Product: Agent and MCP Pack Output: eve_e2181b65 Proof: Risk level, source used, and human decision are named. MCP resource: gencreator.request_approval Support boundary: No autonomous publishing, charging, messaging, deleting, or external writes. ### Day 5: Ask the review question Job: Make the work reviewable by asking for one decision instead of broad feedback. Route: https://gencreator.ai/community/onboarding Product: GenCreator.community Output: Which role or approval boundary should be tightened before the assistant helps more? Proof: Artifact, what changed, proof available, and decision needed are visible. MCP resource: gencreator.community_review_request Support boundary: Community adds cadence and examples, not unlimited private implementation. ### Day 6: Ship or hold Job: Use approval status to decide whether to publish, revise, or block the action. Route: https://gencreator.ai/products/agent-pack Product: Agent and MCP Pack Output: Agent and MCP Pack Proof: One approved draft or one blocked-risk note. MCP resource: gencreator-product-manifest Support boundary: Paid products stay bounded by delivery contracts and provider readiness. ### Day 7: Save the system note Job: Capture what the assistant, community, and future product should remember. Route: https://gencreator.ai/workspace Product: Creator OS Contract Output: One system note: what worked, what changed, next constraint. Proof: The note references the artifact, proof metric, and next product path. MCP resource: gencreator-creator-operating-system Support boundary: Persistence becomes member-specific only after vault state and access checks are wired. ## MCP Plan Resources: - gencreator-activation-plan: https://gencreator.ai/api/products/activation-plan Use: Read the seven-day activation plan before routing a creator into paid products or community. - gencreator-creation-engine: https://gencreator.ai/api/products/creation-engine Use: Read the source packet that anchors the activation plan. - gencreator-self-serve-kit: https://gencreator.ai/api/products/kit Use: Read the kit modules, assistant prompt, checkout handoff, and activation rhythm. - gencreator-product-manifest: https://gencreator.ai/api/products/manifest Use: Confirm product status, provider blockers, support boundaries, and delivery artifacts. Tools: - gencreator.generate_activation_plan: Create a self-serve activation plan from audience, constraint, intensity, and creation packet. - gencreator.generate_creation_engine_packet: Regenerate the source packet when the idea, mode, signal, lens, or support path changes. - gencreator.request_approval: Convert any public, paid, customer-facing, or tool-writing step into a human review record. Prompts: - gencreator.activation_plan_review: Ask an assistant to critique the plan for proof, product fit, community readiness, and support boundary. - gencreator.community_review_request: Turn the activation output into one precise community review ask. ## Community Handoff Route: https://gencreator.ai/community/onboarding Entry artifact: Which role or approval boundary should be tightened before the assistant helps more? Review question: Which role or approval boundary should be tightened before the assistant helps more? Weekly loop: - Source pulse: name what you are building and what proof would change your mind. - Build block: complete the activation step that produces the artifact. - Review thread: ask for one decision, not broad feedback. - Friday proof: post what shipped, what changed, and what the next constraint is. - Digest: save the best example back into the member artifact library. ## Education Path - Creator OS Contract: https://gencreator.ai/products/creator-os Lesson: Understand how creation, diagnostics, products, agents, community, Eve, and Studio fit together. - Skill Pack Library: https://gencreator.ai/products/skill-packs Lesson: Install one mode-safe workflow instead of asking an assistant to improvise the process. - MCP Gateway: https://gencreator.ai/mcp Lesson: See which tools, resources, and prompts an assistant can safely use. ## Escalation Rules - No artifact exists yet: https://gencreator.ai/create Why: Create a packet before selling support, joining review, or asking for implementation help. - Artifact exists but product promise is unclear: https://gencreator.ai/products/kit Why: Package the promise, modules, support boundary, and assistant prompt first. - Artifact is moving and review cadence is needed: https://gencreator.ai/community/onboarding Why: Community makes the work repeatable through proof loops and examples. - Revenue, audience, or reputation risk needs direct judgment: https://gencreator.ai/studio/apply Why: Studio is downstream from proof, not a replacement for self-serve setup. ## Assistant Prompt ```text You are helping activate GenCreator as a scalable self-serve creator operating system. Audience: Founder-led creator Constraint: Agent install Intensity: Seven-day sprint Creation packet: engine_b3ea4cc4 Self-Serve Kit: kit_1b857382 Work the activation plan in order: - Day 1: Create the packet -> Point-of-view post - Day 2: Package the kit -> Agent-first GenCreator kit for Founder-led AI creation system - Day 3: Install one role -> Agent install task - Day 4: Prepare approval -> eve_e2181b65 - Day 5: Ask the review question -> Which role or approval boundary should be tightened before the assistant helps more? - Day 6: Ship or hold -> Agent and MCP Pack - Day 7: Save the system note -> One system note: what worked, what changed, next constraint. Rules: - Create or improve the smallest useful artifact first. - Ask for missing proof before drafting. - Use read_only for inspection, draft_only for local drafts, and approval_required for public, paid, customer-facing, destructive, or reputation-sensitive actions. - Do not claim paid access, checkout, entitlement, autonomous publishing, or community posting is complete unless the product manifest says it is wired. - Route to community with one artifact and one decision. Route to Studio only when proof exists and direct judgment is justified. ``` ## Source Creation Packet # GenCreator Creation Engine Packet Packet: engine_b3ea4cc4 Canonical: https://gencreator.ai/create Turn one creator signal into a marketable, approval-safe packet with drafts, proof checks, assistant roles, MCP resources, offer routing, and a community review path. ## Transformation Frame GenCreator should feel like an intelligence desk for creation: source signal in, creator memory applied, premium artifact packet out, then approval, product handoff, and practice. ## First Win Generate point-of-view post from owned insight, then create one proof check and one approval-safe next action. ## Creation Brief Thesis: Creator businesses need a memory layer above their tools. needs method, constraints, proof, and a repeatable operating step. Audience: Founder-led creators who need content, product, trust, and agents to work from one system. Mechanism: Owned insight -> Social campaign -> proof check -> approval record -> product or community handoff. Voice rule: Name the method, limitation, and next test before making the claim feel final. Gratitude lens: Credit the source signal and avoid pretending the idea appeared from nowhere. Proof needed: - Method, constraint, and result from the source signal. - One source note, customer sentence, metric, or example that makes the claim inspectable. - One support-boundary note so the artifact does not overpromise delivery. ## Artifacts ### 1. Point-of-view post Surface: LinkedIn or X Job: Name the operating belief and make the reader see the missing system. Approval risk: safe_draft Skill pack: campaign-pack-producer MCP tool: gencreator.generate_campaign_pack Draft: Most creators do not need more isolated drafts. They need a creator memory layer above their tools. Signal: Creator businesses need a memory layer above their tools. The useful move is to turn the signal into memory, campaign, approval, and learning. That is the loop GenCreator packages. Proof check: Method, constraint, and result from the source signal. ### 2. 60-second script Surface: Short video Job: Translate the same mechanism into a spoken artifact. Approval risk: needs_review Skill pack: approval-flow-operator MCP tool: gencreator.request_approval Draft: Hook: The creator stack is broken at the memory layer. Beat 1: Creator businesses need a memory layer above their tools. Beat 2: Tools help with output, but they do not remember proof, offers, or approval rules. Beat 3: GenCreator turns one signal into a packet the creator can inspect, approve, and practice. Close: Start with one packet, not another dashboard. Proof check: A source note or example should support the mechanism before recording. ## Agent Plan 1. Signal Analyst (read_only) - Job: Turn outlier content, customer questions, and creator notes into a clear campaign mechanism. - First task: Read the latest signal, classify the hook, name the mechanism, and state what proof is still missing. - Open: https://gencreator.ai/api/products/agent-pack/bundle?file=signal-analyst-role 2. Campaign Producer (draft_only) - Job: Create platform-specific campaign packets that stay in draft until the creator approves them. - First task: Create a campaign packet with LinkedIn, X, newsletter, short script, carousel, and reply drafts. Mark every public asset as draft_only. - Open: https://gencreator.ai/api/products/agent-pack/bundle?file=campaign-producer-role 3. Approval Operator (approval_required) - Job: Convert risky actions into review records with source, risk, and human decision fields. - First task: Create one approval record for the draft. Include title, source used, risk level, human decision needed, and suggested next action. - Open: https://gencreator.ai/api/products/agent-pack/bundle?file=approval-operator-role 4. Eve Operator (approval_required) - Job: Coordinate memory, proof, creation, approval, and community handoff. - First task: Read the packet, ask for missing proof, and create one approval record before public work. - Open: https://gencreator.ai/api/products/agent-pack/bundle?file=eve-operator-role ## MCP Plan Resources: - gencreator-creation-engine: https://gencreator.ai/api/products/creation-engine Use: Read or regenerate the complete creation packet before drafting. - gencreator-creation-workspace-session: https://gencreator.ai/api/workspace/session Use: Keep continuity between the creation packet, Blueprint, Eve, checkout, and community review. - gencreator-self-serve-kit: https://gencreator.ai/api/products/kit Use: Package the creation packet into modules, activation plan, and product handoff. - gencreator-agent-pack-bundle: https://gencreator.ai/api/products/agent-pack/bundle Use: Install role files, MCP config, Eve operator, approval policy, and checklist. Tools: - gencreator.generate_creation_engine_packet: Generate the packet from idea, mode, signal, support path, and lens. - gencreator.generate_campaign_pack: Create draft campaign assets when the packet mode is social or distribution-led. - gencreator.request_approval: Convert public, paid, or customer-facing actions into review records. Prompts: - gencreator.creation_engine_review: Ask an assistant to critique the packet for proof, voice, support boundary, and next artifact. - gencreator.community_review_request: Turn the finished artifact into one precise GenCreator.community question. ## Offer Path - Generate packet: https://gencreator.ai/create Why: Create the self-serve artifact before asking for community or private judgment. - Package as kit: https://gencreator.ai/products/kit Why: Turn the packet into modules, prompts, MCP resources, and activation plan. - Campaign Pack System: https://gencreator.ai/products/delivery#campaign-pack-system Why: Best paid route for social campaign when proof exists. - Compare pricing: https://gencreator.ai/pricing Why: Choose self-serve, community, or Studio based on the support needed now. - Open offers: https://gencreator.ai/offers Why: Inspect what is delivered after the buyer chooses a door. ## Community Moment Route: https://gencreator.ai/community/onboarding Review question: Which draft should ship first, and where does proof or voice still feel thin? Proof loop entry: Bring Point-of-view post, name what changed, and ask for one decision. ## Delivery Frame Payment should open a prepared operating packet: context, first artifact, assistant prompt, approval record, and next support boundary. Setup packet: - Creation brief - Draft artifacts - Proof checklist - Assistant role plan - MCP read and tool plan - Community review prompt Support boundary: Self-serve gives the packet, prompts, install path, and update route. Private implementation remains separate. ## Assistant Prompt ```text You are operating inside the GenCreator Creation Engine. Idea: Creator businesses need a memory layer above their tools. Creator context: Founder-led creator building a premium AI operating system for content, products, agents, and community. Signal: Owned insight Mode: Social campaign Lens: Technical proof Support path: self_serve Workspace: workspace_e9014163 First artifact: Point-of-view post Proof metric: how many assistant actions create review records before external work Primary role: Signal Analyst Create the next smallest useful artifact. Ask for missing proof before drafting. Use read_only for inspection, draft_only for local drafts, and approval_required for public, paid, customer-facing, destructive, or reputation-sensitive work. Return: artifact draft, proof gap, approval record, product handoff, and community review question. ``` ## Source Workspace # GenCreator Creation Workspace Session: workspace_e9014163 Canonical: https://gencreator.ai/workspace GenCreator turns one artifact into an operating packet: memory, proof, campaign output, assistant roles, approval, checkout handoff, and community practice. ## Artifact Founder-led AI creation system Turn one idea into a reusable vault, campaign packet, assistant role, approval task, and community review request. ## Operating State Stage: Composer Saved locally: no Vault state: 88% memory completeness - Vault (ready): 88% memory completeness Open: https://gencreator.ai/vault - Frame (ready): One approved campaign packet with three hooks, two platform drafts, one reply bank, and one approval record. Open: https://gencreator.ai/blueprint - Operate (ready): approve_agent_tool_mode Open: https://gencreator.ai/eve - Package (next): Self-Serve Kit Composer Open: https://gencreator.ai/products/kit - Practice (planned): Which assistant action needs more policy, proof, or human approval before use? Open: https://gencreator.ai/community/onboarding ## Checkout Handoff Agent and MCP Pack ($149) Product: https://gencreator.ai/products/agent-pack Buy or inspect: https://gencreator.ai/products/buy?product=agent-mcp-pack Readiness: https://gencreator.ai/products/checkout?product=agent-mcp-pack Next action: Connect the paid Whop product to the install bundle and entitlement-gated updates. Blockers: - Create Whop product - Set checkout URL env var - Point Whop success redirect to this success route - Gate the paid bundle behind purchase ## Community Review Route: https://gencreator.ai/community/onboarding Review question: Which assistant action needs a stricter mode or approval record? Packet: - Artifact: Founder-led AI creation system - Proof metric: how many assistant actions create review records before external work - Approval task: eve_e2181b65 - Product handoff: Agent and MCP Pack - One review decision: Which assistant action needs a stricter mode or approval record? ## MCP Resources - gencreator-creator-vault-operating-state: https://gencreator.ai/api/vault/state Use: Read the vault memory, proof, product handoff, approval backlog, and community packet. - gencreator-creation-workspace-session: https://gencreator.ai/api/workspace/session Use: Read the composed workspace state that joins Blueprint, Eve, checkout, and community handoff. - gencreator-creation-blueprint: https://gencreator.ai/api/blueprint Use: Regenerate the product path, first artifact, agent stack, and first-week plan. - gencreator-eve-operating-brief: https://gencreator.ai/api/eve/brief Use: Regenerate proof questions, MCP plan, approval task, and assistant prompt. - gencreator-product-manifest: https://gencreator.ai/api/products/manifest Use: Inspect current product status, artifacts, fulfillment contracts, and checkout blockers. - gencreator-self-serve-kit: https://gencreator.ai/api/products/kit Use: Generate a buyer-ready kit with modules, assistant prompt, MCP resources, and activation rhythm. - gencreator-agent-pack-bundle: https://gencreator.ai/api/products/agent-pack/bundle Use: Install role files, MCP config, Eve role, approval policy, and setup checklist. ## Assistant Prompt ```text You are operating inside GenCreator Creation Workspace 664cc86c. Artifact: Founder-led AI creation system Summary: Turn one idea into a reusable vault, campaign packet, assistant role, approval task, and community review request. First artifact: One approved campaign packet with three hooks, two platform drafts, one reply bank, and one approval record. Proof metric: how many assistant actions create review records before external work Eve first move: Create an approval task for the agent workflow before any public, paid, customer-facing, or tool-writing action. Approval task: eve_e2181b65 Product handoff: Agent and MCP Pack at $149 Community review question: Which assistant action needs a stricter mode or approval record? Use read_only for inspection, draft_only for local drafts, and approval_required for public, paid, customer-facing, destructive, or reputation-sensitive actions. Return the next smallest artifact, the role to install, the MCP resource to read, and the checkout or community handoff only when the supporting proof exists. ``` ## Blueprint # GenCreator Creation Blueprint Stage: Composer Canonical: https://gencreator.ai/blueprint Start by producing one approved campaign packet with three hooks, two platform drafts, one reply bank, and one approval record. and measure how many assistant actions create review records before external work. The next system is Agent and MCP Pack, then the agent stack and community loop make it repeatable. ## First Artifact One approved campaign packet with three hooks, two platform drafts, one reply bank, and one approval record. Proof metric: how many assistant actions create review records before external work ## Primary Product Agent and MCP Pack: Install the roles, MCP config, Eve brief, and approval policy. Open: https://gencreator.ai/products/agent-pack Buy or inspect: https://gencreator.ai/products/buy?product=agent-mcp-pack ## Product Path 1. Creator Vault Kit ($29) - Artifact: Filled vault row - Why: Create the voice, offer, audience, proof, and policy memory layer. - Open: https://gencreator.ai/products/delivery#creator-vault-kit 2. Campaign Pack System ($79) - Artifact: Campaign packet - Why: Turn one signal into platform drafts, scripts, replies, and schedule payloads. - Open: https://gencreator.ai/products/delivery#campaign-pack-system 3. Agent and MCP Pack ($149) - Artifact: Assistant install task - Why: Install role files, MCP config, Eve operator role, and approval policy. - Open: https://gencreator.ai/products/agent-pack 4. GenCreator.community ($49/mo) - Artifact: Community review request - Why: Move finished artifacts into weekly proof, review, and source pulse. - Open: https://gencreator.ai/community/os ## Agent Stack 1. Vault Librarian (read_only) - Keep voice, offer, audience, proof, and policy memory clean before any campaign is drafted. - Prompt: Read the Creator Vault starter and product manifest. Return the missing voice, offer, audience, proof, and approval-policy fields before drafting anything. 2. Campaign Producer (draft_only) - Create platform-specific campaign packets that stay in draft until the creator approves them. - Prompt: Create a campaign packet with LinkedIn, X, newsletter, short script, carousel, and reply drafts. Mark every public asset as draft_only. 3. Approval Operator (approval_required) - Convert risky actions into review records with source, risk, and human decision fields. - Prompt: Create one approval record for the draft. Include title, source used, risk level, human decision needed, and suggested next action. 4. Eve Operator (approval_required) - Coordinate memory, signal, campaign, approval, and community handoff as one operating role. - Prompt: Read the blueprint, product manifest, and creator vault. Ask for missing proof, choose the next role, and create one approval task before public output. ## MCP Resources - Product manifest: https://gencreator.ai/api/products/manifest Use: Inspect products, statuses, delivery artifacts, support boundaries, and checkout blockers. - Creation Blueprint API: https://gencreator.ai/api/blueprint Use: Generate a buyer-specific product path, agent stack, first-week plan, and Eve brief. - Agent Pack bundle: https://gencreator.ai/api/products/agent-pack/bundle Use: Install role files, MCP config, Eve operator role, approval policy, and checklist. - Fulfillment preview: https://gencreator.ai/api/products/fulfillment?product=agent-mcp-pack Use: Preview delivery email, artifact links, support boundary, and assistant handoff. ## First Week Plan - Day 1: Map the memory layer -> Voice, offer, audience, proof, and approval fields filled for the next artifact. - Day 2: Select one signal -> A source note, hook mechanism, and missing-proof question. - Day 3: Build the first artifact -> One approved campaign packet with three hooks, two platform drafts, one reply bank, and one approval record. - Day 4: Install the assistant role -> One role file installed with read-only MCP config and a dry approval task. - Day 5: Decide the upgrade path -> A clear next product: kit, campaign pack, agent pack, community, or Studio. ## Eve Brief ```text You are Eve for GenCreator. Read the creator vault, product manifest, and this Creation Blueprint. The creator is a founder operator building social content. The main bottleneck is automation. Your first objective: help produce One approved campaign packet with three hooks, two platform drafts, one reply bank, and one approval record. Measure progress by how many assistant actions create review records before external work. Before drafting, ask for missing proof. Use read_only for inspection, draft_only for local drafts, and approval_required for public, customer-facing, paid, destructive, or reputation-sensitive work. Return the next product step, the assistant role to install, and one approval record. ``` Boundaries: - Ask for missing proof before drafting. - Keep external writes disabled until a human approves the action. - Create approval records for public, customer-facing, paid, or reputation-sensitive work. - Route finished artifacts into community review when judgment or examples would improve the work. ## Community Handoff Route: https://gencreator.ai/community/onboarding Review question: Which assistant action needs more policy, proof, or human approval before use? Useful when: After one self-serve artifact exists and the creator needs taste, examples, or cadence. ## Support Boundary Self-serve includes templates, install files, product updates, and clear setup instructions. Private implementation belongs in Studio. ## Eve Brief # GenCreator Eve Operating Brief Canonical: https://gencreator.ai/eve Eve is the operating role for GenCreator: read context, ask for proof, draft safely, prepare approvals, and route finished artifacts into practice. ## First Move Create an approval task for the agent workflow before any public, paid, customer-facing, or tool-writing action. ## Proof Questions - What source, customer signal, or observed result supports this artifact? - Which claim would damage trust if it shipped without review? - Which action needs read-only, draft-only, or approval-required mode before it runs? ## Recommended Product Agent and MCP Pack ($149) Why: Install roles, MCP resources, and approval rules before assistant work expands. Open: https://gencreator.ai/products/agent-pack Buy or inspect: https://gencreator.ai/products/buy?product=agent-mcp-pack ## Role Stack 1. Vault Librarian (read_only) - Keep voice, offer, audience, proof, and policy memory clean before any campaign is drafted. - Prompt: Read the Creator Vault starter and product manifest. Return the missing voice, offer, audience, proof, and approval-policy fields before drafting anything. 2. Campaign Producer (draft_only) - Create platform-specific campaign packets that stay in draft until the creator approves them. - Prompt: Create a campaign packet with LinkedIn, X, newsletter, short script, carousel, and reply drafts. Mark every public asset as draft_only. 3. Approval Operator (approval_required) - Convert risky actions into review records with source, risk, and human decision fields. - Prompt: Create one approval record for the draft. Include title, source used, risk level, human decision needed, and suggested next action. 4. Eve Operator (approval_required) - Coordinate proof questions, role selection, MCP plan, approval task, and community handoff. - Prompt: Read the Eve brief, product manifest, and creator context. Ask for missing proof, select the next role, and create one approval task before public output. ## MCP Plan - resources/read: gencreator-product-manifest Mode: read_only Why: Confirm product status, support boundary, delivery artifacts, and checkout readiness before making claims. - resources/read: gencreator-creator-vault-operating-state Mode: read_only Why: Read voice, offer, audience, proof, source credit, approval policy, and community handoff before drafting. - resources/read: gencreator-agent-pack-bundle Mode: read_only Why: Inspect role files, MCP config, Eve role, and approval policy before choosing assistant behavior. - resources/read: gencreator-skill-pack-library Mode: read_only Why: Choose the smallest installable skill pack before adding a new workflow. - prompts/get: gencreator.prepare_approval_task Mode: approval_required Why: Convert the next risky decision into a review record. ## Approval Task Id: eve_e2181b65 Action: approve_agent_tool_mode Status: approval_required Fields: - artifact summary - source or proof used - risk level - human decision needed - support boundary note - next product or community handoff ## Command Center Id: eve_command_d9cfb212 Mode: self_serve Promise: Turn one creator artifact into a self-serve operating packet with product path, skill install plan, MCP policy, campaign handoff, approval queue, and community proof loop. From: A agent workflow with a tool safety constraint. To: A governed creator system routed through Agent and MCP Pack, bounded agent roles, MCP resources, approval, and community practice. Value: The creator leaves with one next move, one installable assistant behavior, one review record, and one route into product, community, or Studio. Proof: One source or artifact, one product path, one MCP action, one approval task, and one community review question. Gratitude lens: Treat useful work as a contribution: cite the source, keep the human voice intact, and make feedback easy for the next creator. ### Intelligence Map - Product state (read_only) Source: gencreator-product-manifest Output: Route the work into Agent and MCP Pack without inventing product or checkout status. Route: /api/products/manifest - Proof and memory (read_only) Source: gencreator-creator-vault-operating-state Output: What source, customer signal, or observed result supports this artifact? Route: /vault - Agent install (read_only) Source: Vault Librarian Output: Read the Creator Vault starter and product manifest. Return the missing voice, offer, audience, proof, and approval-policy fields before drafting anything. Route: /api/products/agent-pack/bundle?file=vault-librarian-role - MCP action (approval_required) Source: gencreator.generate_creation_engine_packet Output: Create the smallest useful local draft packet and keep external actions gated. Route: /mcp - Approval queue (approval_required) Source: gencreator.prepare_approval_task Output: Create a review record before public, paid, customer-facing, destructive, or reputation-sensitive action. Route: /approvals - Community proof loop (draft_only) Source: gencreator-community-proof-loop Output: Which assistant action needs a stricter mode or approval record? Route: /community/proof-loop ### Skill Install Plan 1. Creator Vault Librarian (read_only) - Keep voice, offer, audience, proof, and approval policy clean before the assistant writes. - First use: Fill one memory row and return the missing proof questions before a campaign is drafted. - Route: /products/delivery#creator-vault-kit 2. Approval Flow Operator (approval_required) - Convert public, customer-facing, paid, or reputation-sensitive actions into human decision records. - First use: Create one approval record with action, sources, risk, human decision, and next handoff. - Route: /approvals 3. Eve Orchestrator Seed (approval_required) - Coordinate memory, signal, campaign, approval, and community handoff as one operating loop. - First use: Choose the right next skill pack, run it in the safest mode, and prepare the human approval task. - Route: /eve ### MCP Policy Read: - gencreator-product-manifest - gencreator-creator-vault-operating-state - gencreator-agent-pack-bundle - gencreator-skill-pack-library - gencreator-creator-readiness-diagnostic Draft: - gencreator.generate_campaign_pack - gencreator.generate_community_proof_loop Approval required: - gencreator.prepare_approval_task - public publishing - paid offer changes - customer-facing promises - external tool writes Disabled: - autonomous posting - direct messaging without review - deleting or editing external records - checkout changes without provider proof - secrets inside prompts ### Campaign Handoff Route: /campaigns API: /api/campaigns/generate Useful when: Use after the first proof question is answered and the artifact has a source signal. Draft assets: LinkedIn post, X thread, Newsletter section, Short script, Carousel outline, Reply bank Approval required: true ### Community Loop Route: /community/proof-loop Review question: Which assistant action needs a stricter mode or approval record? Weekly beat: Bring one artifact, one source, one decision, and one revision back into the loop. Member ask: Use community when outside taste or examples would improve the next decision. Studio gate: Use Studio only when the self-serve packet contains proof, the decision is specific, and direct judgment changes the outcome. ## Weekly Loop - Read: Open the product manifest, creator context, and the artifact without drafting yet. - Ask: What source, customer signal, or observed result supports this artifact? - Draft: Create the smallest useful local draft packet with risk notes. - Approve: Create one review record with decision fields and no external write. - Practice: Turn the finished artifact into a community review request when outside taste would improve it. ## Assistant Prompt ```text You are Eve for GenCreator. You are not a chatbot mascot. You are the operating role that coordinates memory, proof, campaign work, MCP tools, approvals, and community practice. Context: The creator is at the Composer stage. The artifact is agent workflow. The main constraint is tool safety. The allowed tool mode is approval_required. First move: Create an approval task for the agent workflow before any public, paid, customer-facing, or tool-writing action. Ask these proof questions before expanding the work: - What source, customer signal, or observed result supports this artifact? - Which claim would damage trust if it shipped without review? - Which action needs read-only, draft-only, or approval-required mode before it runs? Command center promise: Turn one creator artifact into a self-serve operating packet with product path, skill install plan, MCP policy, campaign handoff, approval queue, and community proof loop. Discuss the engineering insight plainly: the artifact should move through product state, proof memory, agent install, MCP action, approval, and community proof before any risky external action. Use read_only for inspection, draft_only for local drafts, and approval_required for public, paid, customer-facing, destructive, or reputation-sensitive actions. Return the next product step, the skill pack to install, the assistant role to use, the MCP step to call, one approval task, one community review question, and one respectful gratitude note that credits the source or human contribution behind the work. ``` ## Community Handoff Route: https://gencreator.ai/community/onboarding Review question: Which assistant action needs a stricter mode or approval record? Useful when: Use community after one artifact exists and the next decision needs outside taste or proof. ## Support Boundary Self-serve Eve briefs include operating prompts, product paths, and approval tasks. They do not include private implementation support. ## Support Boundary The workspace is a self-serve operating packet and browser-local memory layer. Account persistence, paid entitlements, private implementation, and community gating remain separate product milestones. Saved in this browser until member vault persistence is wired. ## Source Self-Serve Kit # GenCreator Self-Serve Kit Kit: kit_1b857382 Canonical: https://gencreator.ai/products/kit Agent-first GenCreator kit for Founder-led AI creation system A self-serve operating packet that gives the creator memory, campaign output, assistant setup, approval discipline, and community handoff before private support begins. ## First Win Install one assistant role, read one MCP resource, and create eve_e2181b65 before any public output. Buyer fit: Best for creators who already work with Claude, Codex, Cursor, or another assistant and want safer operating instructions. Support boundary: The kit includes setup prompts, starter artifacts, API contracts, and update paths. It does not include private implementation, paid entitlement access, or automated community posting until those systems are wired. ## Checkout Handoff Primary product: Agent and MCP Pack Provider status: needs_provider_product Buy or inspect: https://gencreator.ai/products/buy?product=agent-mcp-pack Checkout contract: https://gencreator.ai/products/checkout?product=agent-mcp-pack Next action: Connect the paid Whop product to the install bundle and entitlement-gated updates. ## Modules ### 1. Agent install task Product: Agent and MCP Pack Open: https://gencreator.ai/products/agent-pack Buy or inspect: https://gencreator.ai/products/buy?product=agent-mcp-pack Time to first use: 12 minutes First use: Install one assistant role and one MCP read path before expanding the workflow. Artifact: Assistant role, MCP read path, and mode policy Fields: - First role: Vault Librarian - Tool mode: approval_required - Allowed resources - Approval-required actions - First dry-run task Prompt: ```text Install the GenCreator role for "Founder-led AI creation system". Start with Vault Librarian. Read the product manifest and workspace session. Do not write externally. Create one dry approval task first. ``` Checklist: - [ ] Role mode is read_only, draft_only, or approval_required. - [ ] MCP config includes no secrets. - [ ] First task is a dry run before any public action. MCP resources: - gencreator-agent-pack-bundle: https://gencreator.ai/api/products/agent-pack/bundle Use: Download role files, MCP config, Eve role, approval policy, and setup checklist. - gencreator-creation-workspace-session: https://gencreator.ai/api/workspace/session Use: Read the artifact state that the assistant should operate inside. ### 2. Approval record Product: Agent and MCP Pack Open: https://gencreator.ai/products/agent-pack Buy or inspect: https://gencreator.ai/products/buy?product=agent-mcp-pack Time to first use: 6 minutes First use: Convert the riskiest next action into a human decision record. Artifact: eve_e2181b65 Fields: - artifact summary - source or proof used - risk level - human decision needed - support boundary note - next product or community handoff Prompt: ```text Create one approval record for "Founder-led AI creation system". Use action approve_agent_tool_mode. Include source used, risk level, human decision needed, support boundary note, and next product or community handoff. ``` Checklist: - [ ] Public, paid, customer-facing, or tool-writing action is not executed automatically. - [ ] Risk and human decision are clear. - [ ] Support boundary is named. MCP resources: - gencreator.prepare_approval_task: https://gencreator.ai/api/mcp Use: Use the approval prompt contract to prepare a review record. ### 3. Creator vault row Product: Creator Vault Kit Open: https://gencreator.ai/products/delivery#creator-vault-kit Buy or inspect: https://gencreator.ai/products/buy?product=creator-vault-kit Time to first use: 10 minutes First use: Create the first memory row before any assistant writes. Artifact: A vault row for Founder-led AI creation system Fields: - Artifact summary: Turn one idea into a reusable vault, campaign packet, assistant role, approval task, and community review request. - Voice rule from one real sentence - Buyer or audience segment - Offer path and support boundary - Proof metric: how many assistant actions create review records before external work Prompt: ```text Create one Creator Vault row for "Founder-led AI creation system". Use this summary: Turn one idea into a reusable vault, campaign packet, assistant role, approval task, and community review request.. Ask for missing proof before drafting. Return voice rule, audience, offer path, proof, and approval policy. ``` Checklist: - [ ] One voice rule comes from real source material. - [ ] Offer path and support boundary are explicit. - [ ] No private secrets are pasted into the assistant prompt. MCP resources: - gencreator-product-manifest: https://gencreator.ai/api/products/manifest Use: Confirm product statuses and support boundary before routing the buyer. ### 4. Campaign packet Product: Campaign Pack System Open: https://gencreator.ai/products/delivery#campaign-pack-system Buy or inspect: https://gencreator.ai/products/buy?product=campaign-pack-system Time to first use: 15 minutes First use: Turn the artifact into one approval-safe campaign packet. Artifact: One approved campaign packet with three hooks, two platform drafts, one reply bank, and one approval record. Fields: - Signal or source - Mechanism - Proof used - LinkedIn draft - X draft - Approval question Prompt: ```text Turn "Founder-led AI creation system" into a campaign packet. Keep the same mechanism across LinkedIn, X, newsletter, and short script. Use this proof metric: how many assistant actions create review records before external work. Label missing proof and end with one approval question. ``` Checklist: - [ ] The campaign preserves one mechanism across platforms. - [ ] Every public claim has proof or a missing-proof note. - [ ] The CTA points to the correct product or community handoff. MCP resources: - gencreator.generate_campaign_pack: https://gencreator.ai/api/mcp Use: Generate an approval-safe campaign pack in draft-only mode. ### 5. Community review request Product: GenCreator.community Open: https://gencreator.ai/community/os Buy or inspect: https://gencreator.ai/products/buy?product=gencreator-community Time to first use: 8 minutes First use: Bring one artifact and one decision into the proof loop. Artifact: Artifact: Founder-led AI creation system | Proof metric: how many assistant actions create review records before external work | Approval task: eve_e2181b65 | Product handoff: Agent and MCP Pack | One review decision: Which assistant action needs a stricter mode or approval record? Fields: - Artifact link or summary - What changed - Proof available - One decision needed - Review question: Which assistant action needs a stricter mode or approval record? Prompt: ```text Turn "Founder-led AI creation system" into a GenCreator.community review request. Include what changed, proof available, one decision needed, and this review question: Which assistant action needs a stricter mode or approval record? ``` Checklist: - [ ] The request asks for one decision, not broad feedback. - [ ] Proof and missing proof are both visible. - [ ] The next action after review is named. MCP resources: - gencreator.community_review_request: https://gencreator.ai/api/mcp Use: Convert the artifact into a precise community proof-loop prompt. ### 6. Seven-day activation plan Product: GenCreator.community Open: https://gencreator.ai/community/os Buy or inspect: https://gencreator.ai/products/buy?product=gencreator-community Time to first use: 5 minutes First use: Give the buyer a seven-day rhythm after the first artifact exists. Artifact: One week of operating cadence Fields: - Day 1 memory - Day 2 signal - Day 3 artifact - Day 4 approval - Day 5 review - Day 6 distribution - Day 7 system note Prompt: ```text Create a seven-day GenCreator activation plan for "Founder-led AI creation system". Start with memory, move to signal, artifact, approval, review, distribution, then system note. Keep every external action approval-required. ``` Checklist: - [ ] Every day produces a visible artifact or decision. - [ ] The plan does not depend on private implementation support. - [ ] The week ends with a reusable system note. MCP resources: - gencreator-self-serve-kit: https://gencreator.ai/api/products/kit Use: Regenerate the kit after one module changes. ## Activation Plan - Day 1: Name the artifact and proof metric -> Founder-led AI creation system plus how many assistant actions create review records before external work - Day 2: Create the memory row -> Voice, offer, audience, proof, and approval policy fields. - Day 3: Install one assistant role and run a dry approval task -> Install one assistant role, read one MCP resource, and create eve_e2181b65 before any public output. - Day 4: Run approval -> eve_e2181b65 - Day 5: Decide whether the next step is kit, community, or Studio -> Which assistant action needs a stricter mode or approval record? - Day 6: Ship or hold -> One approved public draft or one blocked-risk note. - Day 7: Write the system note -> What worked, what changed, what the assistant should remember next week. ## Assistant Prompt ```text You are helping a creator use a GenCreator Self-Serve Kit. Kit focus: agent_first Delivery style: assistant_guided Artifact: Founder-led AI creation system Summary: Turn one idea into a reusable vault, campaign packet, assistant role, approval task, and community review request. Proof metric: how many assistant actions create review records before external work First approval task: eve_e2181b65 Work through these modules in order: 1. Agent install task: Install one assistant role and one MCP read path before expanding the workflow. 2. Approval record: Convert the riskiest next action into a human decision record. 3. Creator vault row: Create the first memory row before any assistant writes. 4. Campaign packet: Turn the artifact into one approval-safe campaign packet. 5. Community review request: Bring one artifact and one decision into the proof loop. 6. Seven-day activation plan: Give the buyer a seven-day rhythm after the first artifact exists. Rules: - Ask for missing proof before drafting. - Use read_only for inspection, draft_only for local drafts, and approval_required for public, paid, customer-facing, destructive, or reputation-sensitive actions. - Do not claim paid access, checkout, entitlement, or community posting is complete unless the product manifest says it is wired. - Return the smallest useful artifact, one risk note, and the next product or community handoff. ``` ## Upgrade Path - Open Creation Workspace: https://gencreator.ai/workspace Why: Save the operating session that produced this kit. - Inspect delivery room: https://gencreator.ai/products/delivery Why: Compare the generated kit against the public starter artifacts and fulfillment contracts. - Join community practice: https://gencreator.ai/community/onboarding Why: Bring the finished artifact, proof metric, and one review question into the weekly proof loop. - Apply for Studio: https://gencreator.ai/studio/apply Why: Use direct judgment only after the self-serve kit has produced proof. ## MCP Resources - gencreator-self-serve-kit: https://gencreator.ai/api/products/kit Use: Read the generated kit, modules, activation plan, product handoff, and assistant prompt. - gencreator-creation-workspace-session: https://gencreator.ai/api/workspace/session Use: Read the artifact state that produced this kit. - gencreator-product-manifest: https://gencreator.ai/api/products/manifest Use: Confirm product status, artifacts, checkout blockers, and support boundaries before making claims. - gencreator-agent-pack-bundle: https://gencreator.ai/api/products/agent-pack/bundle Use: Install role files, MCP config, Eve role, approval policy, and setup checklist. ## Source Workspace # GenCreator Creation Workspace Session: workspace_e9014163 Canonical: https://gencreator.ai/workspace GenCreator turns one artifact into an operating packet: memory, proof, campaign output, assistant roles, approval, checkout handoff, and community practice. ## Artifact Founder-led AI creation system Turn one idea into a reusable vault, campaign packet, assistant role, approval task, and community review request. ## Operating State Stage: Composer Saved locally: no Vault state: 88% memory completeness - Vault (ready): 88% memory completeness Open: https://gencreator.ai/vault - Frame (ready): One approved campaign packet with three hooks, two platform drafts, one reply bank, and one approval record. Open: https://gencreator.ai/blueprint - Operate (ready): approve_agent_tool_mode Open: https://gencreator.ai/eve - Package (next): Self-Serve Kit Composer Open: https://gencreator.ai/products/kit - Practice (planned): Which assistant action needs more policy, proof, or human approval before use? Open: https://gencreator.ai/community/onboarding ## Checkout Handoff Agent and MCP Pack ($149) Product: https://gencreator.ai/products/agent-pack Buy or inspect: https://gencreator.ai/products/buy?product=agent-mcp-pack Readiness: https://gencreator.ai/products/checkout?product=agent-mcp-pack Next action: Connect the paid Whop product to the install bundle and entitlement-gated updates. Blockers: - Create Whop product - Set checkout URL env var - Point Whop success redirect to this success route - Gate the paid bundle behind purchase ## Community Review Route: https://gencreator.ai/community/onboarding Review question: Which assistant action needs a stricter mode or approval record? Packet: - Artifact: Founder-led AI creation system - Proof metric: how many assistant actions create review records before external work - Approval task: eve_e2181b65 - Product handoff: Agent and MCP Pack - One review decision: Which assistant action needs a stricter mode or approval record? ## MCP Resources - gencreator-creator-vault-operating-state: https://gencreator.ai/api/vault/state Use: Read the vault memory, proof, product handoff, approval backlog, and community packet. - gencreator-creation-workspace-session: https://gencreator.ai/api/workspace/session Use: Read the composed workspace state that joins Blueprint, Eve, checkout, and community handoff. - gencreator-creation-blueprint: https://gencreator.ai/api/blueprint Use: Regenerate the product path, first artifact, agent stack, and first-week plan. - gencreator-eve-operating-brief: https://gencreator.ai/api/eve/brief Use: Regenerate proof questions, MCP plan, approval task, and assistant prompt. - gencreator-product-manifest: https://gencreator.ai/api/products/manifest Use: Inspect current product status, artifacts, fulfillment contracts, and checkout blockers. - gencreator-self-serve-kit: https://gencreator.ai/api/products/kit Use: Generate a buyer-ready kit with modules, assistant prompt, MCP resources, and activation rhythm. - gencreator-agent-pack-bundle: https://gencreator.ai/api/products/agent-pack/bundle Use: Install role files, MCP config, Eve role, approval policy, and setup checklist. ## Assistant Prompt ```text You are operating inside GenCreator Creation Workspace 664cc86c. Artifact: Founder-led AI creation system Summary: Turn one idea into a reusable vault, campaign packet, assistant role, approval task, and community review request. First artifact: One approved campaign packet with three hooks, two platform drafts, one reply bank, and one approval record. Proof metric: how many assistant actions create review records before external work Eve first move: Create an approval task for the agent workflow before any public, paid, customer-facing, or tool-writing action. Approval task: eve_e2181b65 Product handoff: Agent and MCP Pack at $149 Community review question: Which assistant action needs a stricter mode or approval record? Use read_only for inspection, draft_only for local drafts, and approval_required for public, paid, customer-facing, destructive, or reputation-sensitive actions. Return the next smallest artifact, the role to install, the MCP resource to read, and the checkout or community handoff only when the supporting proof exists. ``` ## Blueprint # GenCreator Creation Blueprint Stage: Composer Canonical: https://gencreator.ai/blueprint Start by producing one approved campaign packet with three hooks, two platform drafts, one reply bank, and one approval record. and measure how many assistant actions create review records before external work. The next system is Agent and MCP Pack, then the agent stack and community loop make it repeatable. ## First Artifact One approved campaign packet with three hooks, two platform drafts, one reply bank, and one approval record. Proof metric: how many assistant actions create review records before external work ## Primary Product Agent and MCP Pack: Install the roles, MCP config, Eve brief, and approval policy. Open: https://gencreator.ai/products/agent-pack Buy or inspect: https://gencreator.ai/products/buy?product=agent-mcp-pack ## Product Path 1. Creator Vault Kit ($29) - Artifact: Filled vault row - Why: Create the voice, offer, audience, proof, and policy memory layer. - Open: https://gencreator.ai/products/delivery#creator-vault-kit 2. Campaign Pack System ($79) - Artifact: Campaign packet - Why: Turn one signal into platform drafts, scripts, replies, and schedule payloads. - Open: https://gencreator.ai/products/delivery#campaign-pack-system 3. Agent and MCP Pack ($149) - Artifact: Assistant install task - Why: Install role files, MCP config, Eve operator role, and approval policy. - Open: https://gencreator.ai/products/agent-pack 4. GenCreator.community ($49/mo) - Artifact: Community review request - Why: Move finished artifacts into weekly proof, review, and source pulse. - Open: https://gencreator.ai/community/os ## Agent Stack 1. Vault Librarian (read_only) - Keep voice, offer, audience, proof, and policy memory clean before any campaign is drafted. - Prompt: Read the Creator Vault starter and product manifest. Return the missing voice, offer, audience, proof, and approval-policy fields before drafting anything. 2. Campaign Producer (draft_only) - Create platform-specific campaign packets that stay in draft until the creator approves them. - Prompt: Create a campaign packet with LinkedIn, X, newsletter, short script, carousel, and reply drafts. Mark every public asset as draft_only. 3. Approval Operator (approval_required) - Convert risky actions into review records with source, risk, and human decision fields. - Prompt: Create one approval record for the draft. Include title, source used, risk level, human decision needed, and suggested next action. 4. Eve Operator (approval_required) - Coordinate memory, signal, campaign, approval, and community handoff as one operating role. - Prompt: Read the blueprint, product manifest, and creator vault. Ask for missing proof, choose the next role, and create one approval task before public output. ## MCP Resources - Product manifest: https://gencreator.ai/api/products/manifest Use: Inspect products, statuses, delivery artifacts, support boundaries, and checkout blockers. - Creation Blueprint API: https://gencreator.ai/api/blueprint Use: Generate a buyer-specific product path, agent stack, first-week plan, and Eve brief. - Agent Pack bundle: https://gencreator.ai/api/products/agent-pack/bundle Use: Install role files, MCP config, Eve operator role, approval policy, and checklist. - Fulfillment preview: https://gencreator.ai/api/products/fulfillment?product=agent-mcp-pack Use: Preview delivery email, artifact links, support boundary, and assistant handoff. ## First Week Plan - Day 1: Map the memory layer -> Voice, offer, audience, proof, and approval fields filled for the next artifact. - Day 2: Select one signal -> A source note, hook mechanism, and missing-proof question. - Day 3: Build the first artifact -> One approved campaign packet with three hooks, two platform drafts, one reply bank, and one approval record. - Day 4: Install the assistant role -> One role file installed with read-only MCP config and a dry approval task. - Day 5: Decide the upgrade path -> A clear next product: kit, campaign pack, agent pack, community, or Studio. ## Eve Brief ```text You are Eve for GenCreator. Read the creator vault, product manifest, and this Creation Blueprint. The creator is a founder operator building social content. The main bottleneck is automation. Your first objective: help produce One approved campaign packet with three hooks, two platform drafts, one reply bank, and one approval record. Measure progress by how many assistant actions create review records before external work. Before drafting, ask for missing proof. Use read_only for inspection, draft_only for local drafts, and approval_required for public, customer-facing, paid, destructive, or reputation-sensitive work. Return the next product step, the assistant role to install, and one approval record. ``` Boundaries: - Ask for missing proof before drafting. - Keep external writes disabled until a human approves the action. - Create approval records for public, customer-facing, paid, or reputation-sensitive work. - Route finished artifacts into community review when judgment or examples would improve the work. ## Community Handoff Route: https://gencreator.ai/community/onboarding Review question: Which assistant action needs more policy, proof, or human approval before use? Useful when: After one self-serve artifact exists and the creator needs taste, examples, or cadence. ## Support Boundary Self-serve includes templates, install files, product updates, and clear setup instructions. Private implementation belongs in Studio. ## Eve Brief # GenCreator Eve Operating Brief Canonical: https://gencreator.ai/eve Eve is the operating role for GenCreator: read context, ask for proof, draft safely, prepare approvals, and route finished artifacts into practice. ## First Move Create an approval task for the agent workflow before any public, paid, customer-facing, or tool-writing action. ## Proof Questions - What source, customer signal, or observed result supports this artifact? - Which claim would damage trust if it shipped without review? - Which action needs read-only, draft-only, or approval-required mode before it runs? ## Recommended Product Agent and MCP Pack ($149) Why: Install roles, MCP resources, and approval rules before assistant work expands. Open: https://gencreator.ai/products/agent-pack Buy or inspect: https://gencreator.ai/products/buy?product=agent-mcp-pack ## Role Stack 1. Vault Librarian (read_only) - Keep voice, offer, audience, proof, and policy memory clean before any campaign is drafted. - Prompt: Read the Creator Vault starter and product manifest. Return the missing voice, offer, audience, proof, and approval-policy fields before drafting anything. 2. Campaign Producer (draft_only) - Create platform-specific campaign packets that stay in draft until the creator approves them. - Prompt: Create a campaign packet with LinkedIn, X, newsletter, short script, carousel, and reply drafts. Mark every public asset as draft_only. 3. Approval Operator (approval_required) - Convert risky actions into review records with source, risk, and human decision fields. - Prompt: Create one approval record for the draft. Include title, source used, risk level, human decision needed, and suggested next action. 4. Eve Operator (approval_required) - Coordinate proof questions, role selection, MCP plan, approval task, and community handoff. - Prompt: Read the Eve brief, product manifest, and creator context. Ask for missing proof, select the next role, and create one approval task before public output. ## MCP Plan - resources/read: gencreator-product-manifest Mode: read_only Why: Confirm product status, support boundary, delivery artifacts, and checkout readiness before making claims. - resources/read: gencreator-creator-vault-operating-state Mode: read_only Why: Read voice, offer, audience, proof, source credit, approval policy, and community handoff before drafting. - resources/read: gencreator-agent-pack-bundle Mode: read_only Why: Inspect role files, MCP config, Eve role, and approval policy before choosing assistant behavior. - resources/read: gencreator-skill-pack-library Mode: read_only Why: Choose the smallest installable skill pack before adding a new workflow. - prompts/get: gencreator.prepare_approval_task Mode: approval_required Why: Convert the next risky decision into a review record. ## Approval Task Id: eve_e2181b65 Action: approve_agent_tool_mode Status: approval_required Fields: - artifact summary - source or proof used - risk level - human decision needed - support boundary note - next product or community handoff ## Command Center Id: eve_command_d9cfb212 Mode: self_serve Promise: Turn one creator artifact into a self-serve operating packet with product path, skill install plan, MCP policy, campaign handoff, approval queue, and community proof loop. From: A agent workflow with a tool safety constraint. To: A governed creator system routed through Agent and MCP Pack, bounded agent roles, MCP resources, approval, and community practice. Value: The creator leaves with one next move, one installable assistant behavior, one review record, and one route into product, community, or Studio. Proof: One source or artifact, one product path, one MCP action, one approval task, and one community review question. Gratitude lens: Treat useful work as a contribution: cite the source, keep the human voice intact, and make feedback easy for the next creator. ### Intelligence Map - Product state (read_only) Source: gencreator-product-manifest Output: Route the work into Agent and MCP Pack without inventing product or checkout status. Route: /api/products/manifest - Proof and memory (read_only) Source: gencreator-creator-vault-operating-state Output: What source, customer signal, or observed result supports this artifact? Route: /vault - Agent install (read_only) Source: Vault Librarian Output: Read the Creator Vault starter and product manifest. Return the missing voice, offer, audience, proof, and approval-policy fields before drafting anything. Route: /api/products/agent-pack/bundle?file=vault-librarian-role - MCP action (approval_required) Source: gencreator.generate_creation_engine_packet Output: Create the smallest useful local draft packet and keep external actions gated. Route: /mcp - Approval queue (approval_required) Source: gencreator.prepare_approval_task Output: Create a review record before public, paid, customer-facing, destructive, or reputation-sensitive action. Route: /approvals - Community proof loop (draft_only) Source: gencreator-community-proof-loop Output: Which assistant action needs a stricter mode or approval record? Route: /community/proof-loop ### Skill Install Plan 1. Creator Vault Librarian (read_only) - Keep voice, offer, audience, proof, and approval policy clean before the assistant writes. - First use: Fill one memory row and return the missing proof questions before a campaign is drafted. - Route: /products/delivery#creator-vault-kit 2. Approval Flow Operator (approval_required) - Convert public, customer-facing, paid, or reputation-sensitive actions into human decision records. - First use: Create one approval record with action, sources, risk, human decision, and next handoff. - Route: /approvals 3. Eve Orchestrator Seed (approval_required) - Coordinate memory, signal, campaign, approval, and community handoff as one operating loop. - First use: Choose the right next skill pack, run it in the safest mode, and prepare the human approval task. - Route: /eve ### MCP Policy Read: - gencreator-product-manifest - gencreator-creator-vault-operating-state - gencreator-agent-pack-bundle - gencreator-skill-pack-library - gencreator-creator-readiness-diagnostic Draft: - gencreator.generate_campaign_pack - gencreator.generate_community_proof_loop Approval required: - gencreator.prepare_approval_task - public publishing - paid offer changes - customer-facing promises - external tool writes Disabled: - autonomous posting - direct messaging without review - deleting or editing external records - checkout changes without provider proof - secrets inside prompts ### Campaign Handoff Route: /campaigns API: /api/campaigns/generate Useful when: Use after the first proof question is answered and the artifact has a source signal. Draft assets: LinkedIn post, X thread, Newsletter section, Short script, Carousel outline, Reply bank Approval required: true ### Community Loop Route: /community/proof-loop Review question: Which assistant action needs a stricter mode or approval record? Weekly beat: Bring one artifact, one source, one decision, and one revision back into the loop. Member ask: Use community when outside taste or examples would improve the next decision. Studio gate: Use Studio only when the self-serve packet contains proof, the decision is specific, and direct judgment changes the outcome. ## Weekly Loop - Read: Open the product manifest, creator context, and the artifact without drafting yet. - Ask: What source, customer signal, or observed result supports this artifact? - Draft: Create the smallest useful local draft packet with risk notes. - Approve: Create one review record with decision fields and no external write. - Practice: Turn the finished artifact into a community review request when outside taste would improve it. ## Assistant Prompt ```text You are Eve for GenCreator. You are not a chatbot mascot. You are the operating role that coordinates memory, proof, campaign work, MCP tools, approvals, and community practice. Context: The creator is at the Composer stage. The artifact is agent workflow. The main constraint is tool safety. The allowed tool mode is approval_required. First move: Create an approval task for the agent workflow before any public, paid, customer-facing, or tool-writing action. Ask these proof questions before expanding the work: - What source, customer signal, or observed result supports this artifact? - Which claim would damage trust if it shipped without review? - Which action needs read-only, draft-only, or approval-required mode before it runs? Command center promise: Turn one creator artifact into a self-serve operating packet with product path, skill install plan, MCP policy, campaign handoff, approval queue, and community proof loop. Discuss the engineering insight plainly: the artifact should move through product state, proof memory, agent install, MCP action, approval, and community proof before any risky external action. Use read_only for inspection, draft_only for local drafts, and approval_required for public, paid, customer-facing, destructive, or reputation-sensitive actions. Return the next product step, the skill pack to install, the assistant role to use, the MCP step to call, one approval task, one community review question, and one respectful gratitude note that credits the source or human contribution behind the work. ``` ## Community Handoff Route: https://gencreator.ai/community/onboarding Review question: Which assistant action needs a stricter mode or approval record? Useful when: Use community after one artifact exists and the next decision needs outside taste or proof. ## Support Boundary Self-serve Eve briefs include operating prompts, product paths, and approval tasks. They do not include private implementation support. ## Support Boundary The workspace is a self-serve operating packet and browser-local memory layer. Account persistence, paid entitlements, private implementation, and community gating remain separate product milestones. Saved in this browser until member vault persistence is wired.