diff --git a/docs/marketing/data/seo-geo-audit-workbook.csv b/docs/marketing/data/seo-geo-audit-workbook.csv
new file mode 100644
index 0000000..52040dc
--- /dev/null
+++ b/docs/marketing/data/seo-geo-audit-workbook.csv
@@ -0,0 +1,49 @@
+audit_id,baseline_id,experiment_id,route_role,canonical_path,target_user,query,query_cluster,surface,task_kind,evidence_fields,success_criteria,forbidden_claims,status,next_action
+source_backed_assets_control_google_search_console,source_backed_assets_control,2026_q2_source_backed_assets_intent_validation,control,/product/github-repo-to-launch-package,product_marketer,GitHub repo to launch package,github_repo_to_launch_package,google_search_console,search_console_query_baseline,window_start|window_end|query|page|country|device|impressions|clicks|ctr|average_position,query_data_collected_and_mapped_to_canonical_page,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,export_14_day_search_console_query_page_metrics
+source_backed_assets_control_chatgpt_search,source_backed_assets_control,2026_q2_source_backed_assets_intent_validation,control,/product/github-repo-to-launch-package,product_marketer,GitHub repo to launch package,github_repo_to_launch_package,chatgpt_search,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+source_backed_assets_control_perplexity,source_backed_assets_control,2026_q2_source_backed_assets_intent_validation,control,/product/github-repo-to-launch-package,product_marketer,GitHub repo to launch package,github_repo_to_launch_package,perplexity,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+source_backed_assets_control_google_ai_overview,source_backed_assets_control,2026_q2_source_backed_assets_intent_validation,control,/product/github-repo-to-launch-package,product_marketer,GitHub repo to launch package,github_repo_to_launch_package,google_ai_overview,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+source_backed_assets_control_gemini,source_backed_assets_control,2026_q2_source_backed_assets_intent_validation,control,/product/github-repo-to-launch-package,product_marketer,GitHub repo to launch package,github_repo_to_launch_package,gemini,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+source_backed_assets_control_claude,source_backed_assets_control,2026_q2_source_backed_assets_intent_validation,control,/product/github-repo-to-launch-package,product_marketer,GitHub repo to launch package,github_repo_to_launch_package,claude,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+source_backed_assets_variant_google_search_console,source_backed_assets_variant,2026_q2_source_backed_assets_intent_validation,variant,/product/source-backed-launch-assets,product_marketer,source backed launch assets,source_backed_launch_assets,google_search_console,search_console_query_baseline,window_start|window_end|query|page|country|device|impressions|clicks|ctr|average_position,query_data_collected_and_mapped_to_canonical_page,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,export_14_day_search_console_query_page_metrics
+source_backed_assets_variant_chatgpt_search,source_backed_assets_variant,2026_q2_source_backed_assets_intent_validation,variant,/product/source-backed-launch-assets,product_marketer,source backed launch assets,source_backed_launch_assets,chatgpt_search,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+source_backed_assets_variant_perplexity,source_backed_assets_variant,2026_q2_source_backed_assets_intent_validation,variant,/product/source-backed-launch-assets,product_marketer,source backed launch assets,source_backed_launch_assets,perplexity,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+source_backed_assets_variant_google_ai_overview,source_backed_assets_variant,2026_q2_source_backed_assets_intent_validation,variant,/product/source-backed-launch-assets,product_marketer,source backed launch assets,source_backed_launch_assets,google_ai_overview,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+source_backed_assets_variant_gemini,source_backed_assets_variant,2026_q2_source_backed_assets_intent_validation,variant,/product/source-backed-launch-assets,product_marketer,source backed launch assets,source_backed_launch_assets,gemini,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+source_backed_assets_variant_claude,source_backed_assets_variant,2026_q2_source_backed_assets_intent_validation,variant,/product/source-backed-launch-assets,product_marketer,source backed launch assets,source_backed_launch_assets,claude,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+readme_cards_control_google_search_console,readme_cards_control,2026_q2_readme_cards_intent_validation,control,/product/github-repo-to-launch-package,design_lead,GitHub repo to launch package,github_repo_to_launch_package,google_search_console,search_console_query_baseline,window_start|window_end|query|page|country|device|impressions|clicks|ctr|average_position,query_data_collected_and_mapped_to_canonical_page,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,export_14_day_search_console_query_page_metrics
+readme_cards_control_chatgpt_search,readme_cards_control,2026_q2_readme_cards_intent_validation,control,/product/github-repo-to-launch-package,design_lead,GitHub repo to launch package,github_repo_to_launch_package,chatgpt_search,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+readme_cards_control_perplexity,readme_cards_control,2026_q2_readme_cards_intent_validation,control,/product/github-repo-to-launch-package,design_lead,GitHub repo to launch package,github_repo_to_launch_package,perplexity,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+readme_cards_control_google_ai_overview,readme_cards_control,2026_q2_readme_cards_intent_validation,control,/product/github-repo-to-launch-package,design_lead,GitHub repo to launch package,github_repo_to_launch_package,google_ai_overview,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+readme_cards_control_gemini,readme_cards_control,2026_q2_readme_cards_intent_validation,control,/product/github-repo-to-launch-package,design_lead,GitHub repo to launch package,github_repo_to_launch_package,gemini,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+readme_cards_control_claude,readme_cards_control,2026_q2_readme_cards_intent_validation,control,/product/github-repo-to-launch-package,design_lead,GitHub repo to launch package,github_repo_to_launch_package,claude,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+readme_cards_variant_google_search_console,readme_cards_variant,2026_q2_readme_cards_intent_validation,variant,/product/readme-marketing-cards,design_lead,README marketing cards,readme_marketing_cards,google_search_console,search_console_query_baseline,window_start|window_end|query|page|country|device|impressions|clicks|ctr|average_position,query_data_collected_and_mapped_to_canonical_page,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,export_14_day_search_console_query_page_metrics
+readme_cards_variant_chatgpt_search,readme_cards_variant,2026_q2_readme_cards_intent_validation,variant,/product/readme-marketing-cards,design_lead,README marketing cards,readme_marketing_cards,chatgpt_search,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+readme_cards_variant_perplexity,readme_cards_variant,2026_q2_readme_cards_intent_validation,variant,/product/readme-marketing-cards,design_lead,README marketing cards,readme_marketing_cards,perplexity,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+readme_cards_variant_google_ai_overview,readme_cards_variant,2026_q2_readme_cards_intent_validation,variant,/product/readme-marketing-cards,design_lead,README marketing cards,readme_marketing_cards,google_ai_overview,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+readme_cards_variant_gemini,readme_cards_variant,2026_q2_readme_cards_intent_validation,variant,/product/readme-marketing-cards,design_lead,README marketing cards,readme_marketing_cards,gemini,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+readme_cards_variant_claude,readme_cards_variant,2026_q2_readme_cards_intent_validation,variant,/product/readme-marketing-cards,design_lead,README marketing cards,readme_marketing_cards,claude,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+cold_start_materials_control_google_search_console,cold_start_materials_control,2026_q2_cold_start_materials_intent_validation,control,/product/github-repo-to-launch-package,ai_project_builder,GitHub repo to launch package,github_repo_to_launch_package,google_search_console,search_console_query_baseline,window_start|window_end|query|page|country|device|impressions|clicks|ctr|average_position,query_data_collected_and_mapped_to_canonical_page,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,export_14_day_search_console_query_page_metrics
+cold_start_materials_control_chatgpt_search,cold_start_materials_control,2026_q2_cold_start_materials_intent_validation,control,/product/github-repo-to-launch-package,ai_project_builder,GitHub repo to launch package,github_repo_to_launch_package,chatgpt_search,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+cold_start_materials_control_perplexity,cold_start_materials_control,2026_q2_cold_start_materials_intent_validation,control,/product/github-repo-to-launch-package,ai_project_builder,GitHub repo to launch package,github_repo_to_launch_package,perplexity,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+cold_start_materials_control_google_ai_overview,cold_start_materials_control,2026_q2_cold_start_materials_intent_validation,control,/product/github-repo-to-launch-package,ai_project_builder,GitHub repo to launch package,github_repo_to_launch_package,google_ai_overview,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+cold_start_materials_control_gemini,cold_start_materials_control,2026_q2_cold_start_materials_intent_validation,control,/product/github-repo-to-launch-package,ai_project_builder,GitHub repo to launch package,github_repo_to_launch_package,gemini,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+cold_start_materials_control_claude,cold_start_materials_control,2026_q2_cold_start_materials_intent_validation,control,/product/github-repo-to-launch-package,ai_project_builder,GitHub repo to launch package,github_repo_to_launch_package,claude,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+cold_start_materials_variant_google_search_console,cold_start_materials_variant,2026_q2_cold_start_materials_intent_validation,variant,/product/cold-start-launch-materials,ai_project_builder,cold start launch materials,cold_start_launch_materials,google_search_console,search_console_query_baseline,window_start|window_end|query|page|country|device|impressions|clicks|ctr|average_position,query_data_collected_and_mapped_to_canonical_page,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,export_14_day_search_console_query_page_metrics
+cold_start_materials_variant_chatgpt_search,cold_start_materials_variant,2026_q2_cold_start_materials_intent_validation,variant,/product/cold-start-launch-materials,ai_project_builder,cold start launch materials,cold_start_launch_materials,chatgpt_search,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+cold_start_materials_variant_perplexity,cold_start_materials_variant,2026_q2_cold_start_materials_intent_validation,variant,/product/cold-start-launch-materials,ai_project_builder,cold start launch materials,cold_start_launch_materials,perplexity,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+cold_start_materials_variant_google_ai_overview,cold_start_materials_variant,2026_q2_cold_start_materials_intent_validation,variant,/product/cold-start-launch-materials,ai_project_builder,cold start launch materials,cold_start_launch_materials,google_ai_overview,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+cold_start_materials_variant_gemini,cold_start_materials_variant,2026_q2_cold_start_materials_intent_validation,variant,/product/cold-start-launch-materials,ai_project_builder,cold start launch materials,cold_start_launch_materials,gemini,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+cold_start_materials_variant_claude,cold_start_materials_variant,2026_q2_cold_start_materials_intent_validation,variant,/product/cold-start-launch-materials,ai_project_builder,cold start launch materials,cold_start_launch_materials,claude,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+launch_materials_map_control_google_search_console,launch_materials_map_control,2026_q2_launch_materials_map_intent_validation,control,/product/cold-start-launch-materials,ai_project_builder,cold start launch materials,cold_start_launch_materials,google_search_console,search_console_query_baseline,window_start|window_end|query|page|country|device|impressions|clicks|ctr|average_position,query_data_collected_and_mapped_to_canonical_page,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,export_14_day_search_console_query_page_metrics
+launch_materials_map_control_chatgpt_search,launch_materials_map_control,2026_q2_launch_materials_map_intent_validation,control,/product/cold-start-launch-materials,ai_project_builder,cold start launch materials,cold_start_launch_materials,chatgpt_search,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+launch_materials_map_control_perplexity,launch_materials_map_control,2026_q2_launch_materials_map_intent_validation,control,/product/cold-start-launch-materials,ai_project_builder,cold start launch materials,cold_start_launch_materials,perplexity,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+launch_materials_map_control_google_ai_overview,launch_materials_map_control,2026_q2_launch_materials_map_intent_validation,control,/product/cold-start-launch-materials,ai_project_builder,cold start launch materials,cold_start_launch_materials,google_ai_overview,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+launch_materials_map_control_gemini,launch_materials_map_control,2026_q2_launch_materials_map_intent_validation,control,/product/cold-start-launch-materials,ai_project_builder,cold start launch materials,cold_start_launch_materials,gemini,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+launch_materials_map_control_claude,launch_materials_map_control,2026_q2_launch_materials_map_intent_validation,control,/product/cold-start-launch-materials,ai_project_builder,cold start launch materials,cold_start_launch_materials,claude,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+launch_materials_map_variant_google_search_console,launch_materials_map_variant,2026_q2_launch_materials_map_intent_validation,variant,/product/github-repo-launch-materials-map,ai_project_builder,GitHub repo launch materials map,github_repo_launch_materials_map,google_search_console,search_console_query_baseline,window_start|window_end|query|page|country|device|impressions|clicks|ctr|average_position,query_data_collected_and_mapped_to_canonical_page,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,export_14_day_search_console_query_page_metrics
+launch_materials_map_variant_chatgpt_search,launch_materials_map_variant,2026_q2_launch_materials_map_intent_validation,variant,/product/github-repo-launch-materials-map,ai_project_builder,GitHub repo launch materials map,github_repo_launch_materials_map,chatgpt_search,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+launch_materials_map_variant_perplexity,launch_materials_map_variant,2026_q2_launch_materials_map_intent_validation,variant,/product/github-repo-launch-materials-map,ai_project_builder,GitHub repo launch materials map,github_repo_launch_materials_map,perplexity,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+launch_materials_map_variant_google_ai_overview,launch_materials_map_variant,2026_q2_launch_materials_map_intent_validation,variant,/product/github-repo-launch-materials-map,ai_project_builder,GitHub repo launch materials map,github_repo_launch_materials_map,google_ai_overview,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+launch_materials_map_variant_gemini,launch_materials_map_variant,2026_q2_launch_materials_map_intent_validation,variant,/product/github-repo-launch-materials-map,ai_project_builder,GitHub repo launch materials map,github_repo_launch_materials_map,gemini,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
+launch_materials_map_variant_claude,launch_materials_map_variant,2026_q2_launch_materials_map_intent_validation,variant,/product/github-repo-launch-materials-map,ai_project_builder,GitHub repo launch materials map,github_repo_launch_materials_map,claude,ai_answer_audit,audit_date|mentioned|cited|source_url|answer_summary|accuracy_status|competitors_cited|forbidden_claims_observed,answer_mentions_or_omits_quickfork_without_forbidden_claims,pricing|rankings|revenue|customer_count|conversion_lift|product_hunt_outcome|guaranteed_launch|validated_ai_citation,pending,run_prompt_and_record_mention_citation_accuracy_and_competitors
diff --git a/docs/marketing/research/2026-06-03-seo-geo-evidence-workbook.md b/docs/marketing/research/2026-06-03-seo-geo-evidence-workbook.md
new file mode 100644
index 0000000..4ceef03
--- /dev/null
+++ b/docs/marketing/research/2026-06-03-seo-geo-evidence-workbook.md
@@ -0,0 +1,91 @@
+# 2026-06-03 SEO/GEO Evidence Workbook Research Note
+
+## Summary
+
+The SEO/GEO evidence workbook turns QuickFork's registered Search/AI baseline prompts into a concrete audit queue. Each baseline row expands into one Google Search Console task and five AI-answer audit tasks across ChatGPT Search, Perplexity, Google AI Overview, Gemini, and Claude.
+
+This is a measurement slice, not SEO/GEO validation.
+
+## Why This Slice Matters
+
+- QuickFork now has multiple published product, use-case, resource, tool, and example routes, but the active experiments still have `pending_evidence`.
+- Search Console exports and AI-answer audits are manual enough that evidence can drift unless the exact route, query, surface, required fields, and forbidden claims are fixed.
+- A supervisor-grade workbook lets growth reviewers audit the same experiment contract across GA4, Search Console, and answer-engine surfaces before choosing the next landing page or product slice.
+- This supports the current P0 priority: collect evidence before scaling more pages.
+
+## Target User
+
+- SEO/GEO supervisor reviewing whether QuickFork is indexable, extractable, and accurately summarized.
+- Growth operator collecting route-level Search Console and AI-answer evidence.
+- Product marketer deciding whether a page promise deserves more content, product, or paid-package investment.
+
+## Lifecycle Position
+
+- Discovery: query clusters and AI prompts are fixed by the Search/AI baseline contract.
+- Validation: workbook rows specify the evidence needed before route comparisons can be decided.
+- Activation: workbook decisions must be read alongside generation, artifact export, and launch materials map copy behavior.
+- Monetization: paid package decisions still require lead quality and willingness-to-pay evidence.
+
+## Growth Contract
+
+Primary metric:
+
+- Completion of Search Console and AI-answer audit rows for each active experiment that requires search or AI evidence.
+
+Supporting metrics:
+
+- `page_view`, `cta_clicked`, `generation_started`, `generation_completed`, and `generation_failed` over the same 14-day window.
+- `launch_materials_map_copied` for `2026_q2_launch_materials_map_intent_validation`.
+- Lead-quality fields for full launch package requests when bottom-funnel pages are involved.
+
+Guardrails:
+
+- Do not treat workbook rows, prompt coverage, or page publication as search demand.
+- Do not claim ranking wins, AI citations, revenue, customer count, conversion lift, Product Hunt results, guaranteed launches, or pricing validation.
+- Keep the workbook PII-free: no raw emails, raw repo content, source notes, private launch notes, tokens, secrets, raw artifact bodies, or customer proof.
+
+## Implemented Surface
+
+- `src/marketing/seo-geo-audit-workbook.ts`
+- `src/marketing/seo-geo-audit-workbook.test.ts`
+- `docs/marketing/data/seo-geo-audit-workbook.csv`
+- `docs/superpowers/plans/2026-06-03-seo-geo-evidence-workbook.md`
+
+## Workbook Contract
+
+Every Search/AI baseline row now produces six audit rows:
+
+- `google_search_console`: collect `window_start`, `window_end`, `query`, `page`, `country`, `device`, `impressions`, `clicks`, `ctr`, and `average_position`.
+- `chatgpt_search`, `perplexity`, `google_ai_overview`, `gemini`, and `claude`: collect `audit_date`, `mentioned`, `cited`, `source_url`, `answer_summary`, `accuracy_status`, `competitors_cited`, and `forbidden_claims_observed`.
+
+The workbook keeps each row tied to:
+
+- Baseline id.
+- Experiment id.
+- Control or variant role.
+- Canonical path.
+- Target user.
+- Query cluster.
+- Surface.
+- Expected terms.
+- Forbidden claims.
+- Pending status and next action.
+
+## Current Verification
+
+- RED test failed first because `src/marketing/seo-geo-audit-workbook.ts` did not exist.
+- GREEN focused verification passed after adding the module and CSV mirror: `npm test -- src/marketing/seo-geo-audit-workbook.test.ts` returned 1 file passed and 5 tests passed.
+- Related verification passed: `npm test -- src/marketing/seo-geo-audit-workbook.test.ts src/marketing/search-ai-baseline.test.ts src/marketing/growth-experiment-report.test.ts` returned 3 files passed and 21 tests passed.
+- Initial full `npm test` attempts with the default forks pool hit Vitest worker startup timeouts in this local environment.
+- Targeted reruns of the affected files passed, including `npm test -- src/App.test.tsx -t "keeps the generator studio|full launch package contact|landing architecture"` and `npm test -- src/server/marketing/launch-package-fit.test.ts`.
+- Full verification passed with threads pool: `npm test -- --testTimeout=30000 --pool=threads` returned 25 files passed and 168 tests passed.
+- Build verification passed: `npm run build` completed TypeScript build and Vite production build.
+- Diff hygiene passed: `git diff --check` returned no whitespace errors.
+
+## Decision
+
+Treat the workbook as P0 measurement infrastructure. It makes SEO/GEO evidence collection executable, but it does not prove indexed demand, answer-engine citation, activation quality, lead quality, or willingness to pay.
+
+## Next Validation Step
+
+Use the workbook to collect 14-day Search Console and AI-answer evidence for the active validation experiments, then fill `docs/marketing/data/growth-experiment-evidence.csv` and decide whether each page should be promoted, rewritten, consolidated, or paused.
diff --git a/docs/marketing/research/2026-06-03-sites-output-enablement.md b/docs/marketing/research/2026-06-03-sites-output-enablement.md
new file mode 100644
index 0000000..887567f
--- /dev/null
+++ b/docs/marketing/research/2026-06-03-sites-output-enablement.md
@@ -0,0 +1,45 @@
+# Sites Output Enablement
+
+Date: 2026-06-03
+
+## Hypothesis
+
+If QuickFork exposes Sites as part of the launch package, founders, open-source maintainers, and DevRel operators can turn the same repository evidence into a reviewable site/page spine instead of treating launch output as only README, deck, social, visual, and outreach files.
+
+## Lifecycle Stage
+
+- Stage: Activation to monetization.
+- Target user: AI/devtool builders, open-source maintainers, technical founders, and DevRel operators preparing a public launch.
+- Product surface: generated launch brief, studio output presets, and full launch package qualification form.
+- CTA: generate a similar launch package or request a reviewed full launch package.
+- Primary metric: `launch_artifact_copied` or `launch_artifact_downloaded` where `artifact_type=site`; `sales_contact_requested` with package scope including `sites`.
+- Guardrail metric: no raw README text, artifact body, email, token, secret, unsupported customer proof, ranking, revenue, or guaranteed-growth claim in browser analytics.
+
+## Change
+
+- Added `site` as a generated launch materials channel.
+- Added a source-backed "Launch site page spine" export artifact.
+- Added a Sites output preset for the studio.
+- Added Sites to full launch package qualification scope.
+- Kept Sites as a reviewable page spine, not automatic public publishing.
+
+## Evidence Observed
+
+- RED generation test failed first because the launch brief did not include the site channel.
+- RED App tests failed first because the studio and contact scope form did not expose Sites.
+- Focused GREEN verification passed after implementation:
+  - `npm test -- src/server/generation/generation.test.ts -t "source-backed free repo launch brief"`
+  - `npm test -- src/App.test.tsx -t "keeps the generator studio|full launch package contact"`
+  - `npm test -- src/server/marketing/launch-package-fit.test.ts src/server/marketing/lead-capture.test.ts -t "launch package fit|full launch package qualification"`
+- Related focused verification passed after the final scope-count alignment:
+  - `npm test -- src/server/generation/generation.test.ts -t "source-backed free repo launch brief"` returned 1 file passed and 1 selected test passed.
+  - `npm test -- src/App.test.tsx -t "keeps the generator studio|full launch package contact|landing architecture"` returned 1 file passed and 3 selected tests passed.
+  - `npm test -- src/server/marketing/launch-package-fit.test.ts src/server/marketing/lead-capture.test.ts -t "launch package fit|full launch package qualification"` returned 2 files passed and 2 selected tests passed.
+- Full verification with the default forks pool hit Vitest worker startup timeout in this local environment, but no Sites assertion failed after the scope-count fix.
+- Full verification passed with threads pool: `npm test -- --testTimeout=30000 --pool=threads` returned 25 files passed and 168 tests passed.
+- Build verification passed: `npm run build` completed TypeScript build and Vite production build.
+- Diff hygiene passed: `git diff --check` returned no whitespace errors.
+
+## Remaining Validation Gap
+
+No production users have copied, downloaded, or requested the Sites scope yet. Treat the output as a monetization hypothesis until artifact interactions, full-package requests, or interviews show that launch site/page support is more valuable than the existing README, social, deck, visual, and outreach outputs.
diff --git a/docs/plans/2026-06-01-quickfork-growth-lifecycle-prioritization.md b/docs/plans/2026-06-01-quickfork-growth-lifecycle-prioritization.md
index 2377481..e98ca66 100644
--- a/docs/plans/2026-06-01-quickfork-growth-lifecycle-prioritization.md
+++ b/docs/plans/2026-06-01-quickfork-growth-lifecycle-prioritization.md
@@ -1646,3 +1646,113 @@ Decision:
 Next action:
 
 - Run full verification, merge, production smoke, then use `2026_q2_launch_materials_map_intent_validation` for 14 days of GA4, Search Console, and AI-answer evidence collection.
+
+## 2026-06-03 Sites Output Enablement Slice
+
+Hypothesis:
+
+- If QuickFork exposes Sites as part of the launch package, founders, open-source maintainers, and DevRel operators can turn repository evidence into a reviewable launch site/page spine instead of treating launch output as only README, deck, social, visual, and outreach material.
+
+Lifecycle stage:
+
+- Activation to monetization.
+
+Target user:
+
+- AI/devtool builders, open-source maintainers, technical founders, and DevRel operators preparing a public launch.
+
+Changed surface:
+
+- `src/server/generation/types.ts`
+- `src/server/generation/launch-brief.ts`
+- `src/server/generation/generation.test.ts`
+- `src/components/landing/HeroSection.tsx`
+- `src/components/studio/ProductStudio.tsx`
+- `src/components/marketing/LeadCaptureForm.tsx`
+- `src/core/pipeline.ts`
+- `src/App.test.tsx`
+- `src/server/marketing/launch-package-fit.test.ts`
+- `src/server/marketing/lead-capture.test.ts`
+- `docs/marketing/research/2026-06-03-sites-output-enablement.md`
+
+Metric:
+
+- `launch_artifact_copied` or `launch_artifact_downloaded` where `artifact_type=site`.
+- `sales_contact_requested` with package scope including `sites`.
+
+Guardrail:
+
+- Sites output is a reviewable page spine, not automatic publishing.
+- Do not send raw README text, raw artifact body, email, token, secret, unsupported customer proof, rankings, revenue, pricing, or guaranteed-growth claims to browser analytics.
+
+Evidence observed:
+
+- RED generation test failed first because the launch brief did not include the site channel.
+- RED App tests failed first because the studio and full launch package form did not expose Sites.
+- Focused generation verification passed: `npm test -- src/server/generation/generation.test.ts -t "source-backed free repo launch brief"` returned 1 file passed and 1 selected test passed.
+- Focused App verification passed: `npm test -- src/App.test.tsx -t "keeps the generator studio|full launch package contact|landing architecture"` returned 1 file passed and 3 selected tests passed.
+- Focused lead qualification verification passed: `npm test -- src/server/marketing/launch-package-fit.test.ts src/server/marketing/lead-capture.test.ts -t "launch package fit|full launch package qualification"` returned 2 files passed and 2 selected tests passed.
+- Full verification passed with threads pool: `npm test -- --testTimeout=30000 --pool=threads` returned 25 files passed and 168 tests passed.
+- Build verification passed: `npm run build` completed TypeScript build and Vite production build.
+- Diff hygiene passed: `git diff --check` returned no whitespace errors.
+
+Decision:
+
+- Treat Sites as a product-output and monetization hypothesis. It fills a landing-page material gap in the launch package, but production users have not yet copied, downloaded, or requested this scope.
+
+Next action:
+
+- Track site artifact copy/download and full launch package requests with `sites` scope before investing in full hosted site generation, CMS export, or public landing-page automation.
+
+## 2026-06-03 SEO/GEO Evidence Workbook Slice
+
+Hypothesis:
+
+- If QuickFork expands each registered Search/AI baseline row into a surface-specific audit workbook, SEO/GEO reviewers can collect Search Console and AI-answer evidence without drifting from the experiment contract.
+
+Lifecycle stage:
+
+- Validation, P0 measurement and evidence.
+
+Target user:
+
+- SEO/GEO supervisor auditing whether QuickFork is indexable, extractable, and accurately summarized.
+- Growth operator collecting Search Console and AI-answer evidence for active page-intent experiments.
+
+Changed surface:
+
+- `src/marketing/seo-geo-audit-workbook.ts`
+- `src/marketing/seo-geo-audit-workbook.test.ts`
+- `docs/marketing/data/seo-geo-audit-workbook.csv`
+- `docs/marketing/research/2026-06-03-seo-geo-evidence-workbook.md`
+- `docs/superpowers/plans/2026-06-03-seo-geo-evidence-workbook.md`
+
+Metric:
+
+- Completion of Search Console and AI-answer workbook rows for experiments that require search or AI evidence.
+- Supporting route metrics remain `page_view`, `cta_clicked`, `generation_started`, `generation_completed`, and `generation_failed` over the same 14-day window.
+
+Guardrail:
+
+- Do not treat a workbook row, prompt, route, or crawler asset as validated search demand or AI visibility.
+- Do not claim ranking wins, AI citations, customer traction, revenue, conversion lift, pricing validation, Product Hunt outcomes, or guaranteed launches.
+- Keep workbook rows free of raw emails, raw repo text, source notes, private launch notes, tokens, secrets, raw artifacts, or unsupported proof.
+
+Evidence observed:
+
+- RED test failed first because the SEO/GEO audit workbook module did not exist.
+- Focused verification passed after adding the module and CSV mirror: `npm test -- src/marketing/seo-geo-audit-workbook.test.ts` returned 1 file passed and 5 tests passed.
+- Related verification passed: `npm test -- src/marketing/seo-geo-audit-workbook.test.ts src/marketing/search-ai-baseline.test.ts src/marketing/growth-experiment-report.test.ts` returned 3 files passed and 21 tests passed.
+- Initial full `npm test` attempts with the default forks pool hit Vitest worker startup timeouts in this local environment.
+- Targeted reruns of the affected files passed, including `npm test -- src/App.test.tsx -t "keeps the generator studio|full launch package contact|landing architecture"` and `npm test -- src/server/marketing/launch-package-fit.test.ts`.
+- Full verification passed with threads pool: `npm test -- --testTimeout=30000 --pool=threads` returned 25 files passed and 168 tests passed.
+- Build verification passed: `npm run build` completed TypeScript build and Vite production build.
+- Diff hygiene passed: `git diff --check` returned no whitespace errors.
+
+Decision:
+
+- Treat this as P0 measurement infrastructure, not a validation result. It makes the current active experiments reviewable by a senior SEO/GEO operator before QuickFork scales more landing pages.
+
+Next action:
+
+- Run related/full verification, then use the workbook to collect 14-day Search Console and AI-answer evidence before deciding which active page-intent experiments should be promoted, rewritten, consolidated, or paused.
diff --git a/docs/superpowers/plans/2026-06-03-seo-geo-evidence-workbook.md b/docs/superpowers/plans/2026-06-03-seo-geo-evidence-workbook.md
new file mode 100644
index 0000000..4f6ed57
--- /dev/null
+++ b/docs/superpowers/plans/2026-06-03-seo-geo-evidence-workbook.md
@@ -0,0 +1,156 @@
+# SEO GEO Evidence Workbook Implementation Plan
+
+> **For agentic workers:** REQUIRED SUB-SKILL: Use superpowers:subagent-driven-development (recommended) or superpowers:executing-plans to implement this plan task-by-task. Steps use checkbox (`- [ ]`) syntax for tracking.
+
+**Goal:** Build a test-covered SEO/GEO evidence workbook that expands QuickFork's registered Search/AI baseline rows into human-auditable Search Console and AI-answer collection tasks.
+
+**Architecture:** Add a focused marketing module that derives audit rows from `search-ai-baseline.ts` rather than duplicating experiment definitions. Add an editable CSV mirror for human audit work and a research note explaining the validation boundary.
+
+**Tech Stack:** TypeScript, Vitest, CSV docs, Markdown docs.
+
+---
+
+### Task 1: SEO/GEO Audit Workbook Tests
+
+**Files:**
+- Create: `src/marketing/seo-geo-audit-workbook.test.ts`
+
+- [ ] **Step 1: Write the failing test**
+
+Create `src/marketing/seo-geo-audit-workbook.test.ts` with tests that import `searchAiBaselineRows`, `searchBaselineSurfaces`, and the future workbook helpers. The tests must assert:
+
+- Every baseline row expands to one workbook row per surface.
+- `google_search_console` rows use task kind `search_console_query_baseline`.
+- ChatGPT Search, Perplexity, Google AI Overview, Gemini, and Claude rows use task kind `ai_answer_audit`.
+- Every row preserves `baselineId`, `experimentId`, `routeRole`, `canonicalPath`, `targetUser`, `query`, `queryCluster`, expected terms, forbidden claims, status `pending`, and a non-empty next action.
+- The editable CSV at `docs/marketing/data/seo-geo-audit-workbook.csv` mirrors the generated workbook.
+- The rendered runbook names the experiment, routes, Search Console, AI surfaces, required evidence fields, and explicitly keeps the decision as `insufficient_data`.
+- The rendered runbook does not claim validated demand, ranking wins, revenue lift, customer traction, or guaranteed launch results.
+
+- [ ] **Step 2: Run test to verify it fails**
+
+Run:
+
+```bash
+npm test -- src/marketing/seo-geo-audit-workbook.test.ts
+```
+
+Expected: FAIL because `src/marketing/seo-geo-audit-workbook.ts` and the CSV do not exist yet.
+
+### Task 2: SEO/GEO Audit Workbook Module
+
+**Files:**
+- Create: `src/marketing/seo-geo-audit-workbook.ts`
+
+- [ ] **Step 1: Implement the minimal module**
+
+Create `src/marketing/seo-geo-audit-workbook.ts` with:
+
+- `seoGeoAuditStatuses = ["pending", "collected", "needs_reaudit"]`.
+- Task kinds `search_console_query_baseline` and `ai_answer_audit`.
+- `buildSeoGeoAuditWorkbookRows()` that maps every `SearchAiBaselineRow` surface to a `SeoGeoAuditWorkbookRow`.
+- `getSeoGeoAuditWorkbookRowsForExperiment(experimentId)`.
+- `renderSeoGeoAuditWorkbook(experimentId)` that returns a Markdown runbook with route rows, evidence fields, guardrails, and the decision line `Decision: insufficient_data_until_14_day_ga4_search_console_and_ai_answer_evidence_exists`.
+
+- [ ] **Step 2: Run focused test**
+
+Run:
+
+```bash
+npm test -- src/marketing/seo-geo-audit-workbook.test.ts
+```
+
+Expected: still FAIL until the editable CSV is added.
+
+### Task 3: Editable SEO/GEO Audit CSV
+
+**Files:**
+- Create: `docs/marketing/data/seo-geo-audit-workbook.csv`
+
+- [ ] **Step 1: Add the CSV mirror**
+
+Create a CSV with this header:
+
+```text
+audit_id,baseline_id,experiment_id,route_role,canonical_path,target_user,query,query_cluster,surface,task_kind,evidence_fields,success_criteria,forbidden_claims,status,next_action
+```
+
+Add one row per baseline/surface pair. Use pipe-delimited lists inside fields. Keep all values PII-free and claim-safe.
+
+- [ ] **Step 2: Run focused test**
+
+Run:
+
+```bash
+npm test -- src/marketing/seo-geo-audit-workbook.test.ts
+```
+
+Expected: PASS.
+
+### Task 4: Research Note And Lifecycle Plan Update
+
+**Files:**
+- Create: `docs/marketing/research/2026-06-03-seo-geo-evidence-workbook.md`
+- Modify: `docs/plans/2026-06-01-quickfork-growth-lifecycle-prioritization.md`
+
+- [ ] **Step 1: Add research note**
+
+Document:
+
+- Hypothesis: a generated audit workbook will reduce evidence drift across GA4, Search Console, and AI-answer audits.
+- Lifecycle stage: Validation.
+- Target user: SEO/GEO supervisor and growth operator.
+- Changed surface: TypeScript workbook, CSV mirror, runbook renderer.
+- Metric: completion of Search Console and AI-answer evidence rows for active experiments.
+- Guardrail: no SEO/GEO, pricing, revenue, customer, or AI-citation claims until evidence is collected.
+- Next action: run the workbook for each active experiment and fill the evidence CSV after the 14-day window.
+
+- [ ] **Step 2: Update lifecycle plan**
+
+Append a short dated section that records the SEO/GEO evidence workbook as a P0 measurement slice and states that it is instrumentation, not validation.
+
+### Task 5: Verification
+
+**Files:**
+- Verify only.
+
+- [ ] **Step 1: Run focused related tests**
+
+Run:
+
+```bash
+npm test -- src/marketing/seo-geo-audit-workbook.test.ts src/marketing/search-ai-baseline.test.ts src/marketing/growth-experiment-report.test.ts
+```
+
+Expected: PASS.
+
+- [ ] **Step 2: Run full verification**
+
+Run:
+
+```bash
+npm test
+npm run build
+git diff --check
+```
+
+Expected: PASS. If the full suite fails for unrelated pre-existing reasons, record the exact failure and keep the focused verification result.
+
+### Task 6: Obsidian Mirror
+
+**Files:**
+- Modify: `/Users/moose/Documents/Moose's Birth ID/QuickFork/SEO Growth/2026-06-01 QuickFork Cold Start Launch Growth Iteration.md`
+
+- [ ] **Step 1: Append dated entry**
+
+Append a dated entry with repo refs, hypothesis, lifecycle stage, target user, changed surface, metric, guardrail, evidence observed, decision, and next action.
+
+- [ ] **Step 2: Verify entry exists**
+
+Run:
+
+```bash
+tail -n 120 "/Users/moose/Documents/Moose's Birth ID/QuickFork/SEO Growth/2026-06-01 QuickFork Cold Start Launch Growth Iteration.md"
+```
+
+Expected: The new SEO/GEO evidence workbook entry is visible and states that the workbook is not validation evidence by itself.
diff --git a/src/App.test.tsx b/src/App.test.tsx
index 20d433f..3f8b97e 100644
--- a/src/App.test.tsx
+++ b/src/App.test.tsx
@@ -106,6 +106,7 @@ describe("App", () => {
     expect(screen.getAllByText(/README/i).length).toBeGreaterThan(0);
     expect(screen.getAllByText(/PPT/i).length).toBeGreaterThan(0);
     expect(screen.getAllByText(/Social/i).length).toBeGreaterThan(0);
+    expect(screen.getAllByText(/Sites/i).length).toBeGreaterThan(0);
   });
 
   it("tracks route-level page views with funnel intent metadata", () => {
@@ -1057,6 +1058,7 @@ describe("App", () => {
     fireEvent.click(screen.getByLabelText(/^Social$/i));
     fireEvent.click(screen.getByLabelText(/^Deck$/i));
     fireEvent.click(screen.getByLabelText(/^Outreach$/i));
+    fireEvent.click(screen.getByLabelText(/^Sites$/i));
     fireEvent.click(screen.getByLabelText(/^Visual explainer$/i));
     fireEvent.click(screen.getByLabelText(/human review needed/i));
     fireEvent.change(screen.getByLabelText(/launch notes/i), {
@@ -1080,7 +1082,7 @@ describe("App", () => {
         launchTimeline: "within_30_days",
         packageModel: "single_launch",
         buyingTrigger: "launch_deadline",
-        packageScope: ["readme", "social", "deck", "outreach", "visual_explainer"],
+        packageScope: ["readme", "social", "deck", "outreach", "sites", "visual_explainer"],
         humanReviewNeeded: true,
         notes: "Launching an AI repo and need source-backed README, deck, and outreach review.",
       },
@@ -1101,7 +1103,7 @@ describe("App", () => {
           launch_timeline: "within_30_days",
           package_model: "single_launch",
           buying_trigger: "launch_deadline",
-          package_scope_count: 5,
+          package_scope_count: 6,
           human_review_needed: true,
         }),
       ]),
diff --git a/src/components/landing/HeroSection.tsx b/src/components/landing/HeroSection.tsx
index 905d33b..297ee44 100644
--- a/src/components/landing/HeroSection.tsx
+++ b/src/components/landing/HeroSection.tsx
@@ -142,7 +142,7 @@ interface RepoLaunchMaterialsMapSummary {
 }
 
 interface RepoLaunchMaterialChannelSummary {
-  type: "readme" | "social" | "deck" | "visual" | "outreach";
+  type: "readme" | "social" | "deck" | "site" | "visual" | "outreach";
   label: string;
   primaryUser: string;
   jobToBeDone: string;
@@ -154,7 +154,7 @@ interface RepoLaunchMaterialChannelSummary {
 }
 
 interface RepoLaunchBriefArtifactSummary {
-  type: "audience" | "story_map" | "materials_map" | "readme" | "social" | "deck" | "outreach" | "visual";
+  type: "audience" | "story_map" | "materials_map" | "readme" | "social" | "deck" | "site" | "outreach" | "visual";
   label: string;
   fileName: string;
   body: string;
@@ -706,7 +706,7 @@ function LaunchBriefPanel({
       <div className="launchPackageCta">
         <div>
           <strong>Need a complete launch package?</strong>
-          <small>Request reviewed README, social, deck, outreach, and visual assets before publishing.</small>
+          <small>Request reviewed README, social, deck, site, outreach, and visual assets before publishing.</small>
         </div>
         <a className="primaryButton" href={fullLaunchPackageHref} onClick={handleFullLaunchPackageIntent}>
           Request full launch package
diff --git a/src/components/marketing/LeadCaptureForm.tsx b/src/components/marketing/LeadCaptureForm.tsx
index c3a7da1..9638c24 100644
--- a/src/components/marketing/LeadCaptureForm.tsx
+++ b/src/components/marketing/LeadCaptureForm.tsx
@@ -302,6 +302,7 @@ const launchPackageScopeOptions = [
   { value: "social", label: "Social" },
   { value: "deck", label: "Deck" },
   { value: "outreach", label: "Outreach" },
+  { value: "sites", label: "Sites" },
   { value: "visual_explainer", label: "Visual explainer" },
 ] as const;
 
diff --git a/src/components/studio/ProductStudio.tsx b/src/components/studio/ProductStudio.tsx
index 47ed40c..1568a39 100644
--- a/src/components/studio/ProductStudio.tsx
+++ b/src/components/studio/ProductStudio.tsx
@@ -152,6 +152,7 @@ export function ProductStudio() {
             <span>README</span>
             <span>PPT</span>
             <span>Social</span>
+            <span>Sites</span>
           </div>
         </div>
       </div>
diff --git a/src/core/pipeline.ts b/src/core/pipeline.ts
index 9f9fd9a..6f8704f 100644
--- a/src/core/pipeline.ts
+++ b/src/core/pipeline.ts
@@ -19,7 +19,7 @@ export interface ModelSettings {
 export interface AssetPreset {
   id: string;
   label: string;
-  useCase: "README" | "PPT" | "Social";
+  useCase: "README" | "PPT" | "Social" | "Sites";
   size: string;
   aspect: string;
   note: string;
@@ -87,6 +87,14 @@ export const ASSET_PRESETS: AssetPreset[] = [
     aspect: "16:9",
     note: "Fits widescreen decks and product updates without manual cropping.",
   },
+  {
+    id: "site-hero",
+    label: "Launch site hero",
+    useCase: "Sites",
+    size: "1600x900",
+    aspect: "16:9",
+    note: "Fits a product launch site hero, showcase page, or campaign landing page preview.",
+  },
   {
     id: "x-linkedin-landscape",
     label: "X / LinkedIn landscape",
diff --git a/src/marketing/seo-geo-audit-workbook.test.ts b/src/marketing/seo-geo-audit-workbook.test.ts
new file mode 100644
index 0000000..151c611
--- /dev/null
+++ b/src/marketing/seo-geo-audit-workbook.test.ts
@@ -0,0 +1,190 @@
+import { readFileSync } from "node:fs";
+import { join } from "node:path";
+
+import { describe, expect, it } from "vitest";
+
+import {
+  aiSearchAuditSurfaces,
+  searchAiBaselineRows,
+  searchBaselineSurfaces,
+  type SearchBaselineSurface,
+} from "./search-ai-baseline";
+import {
+  buildSeoGeoAuditWorkbookRows,
+  getSeoGeoAuditWorkbookRowsForExperiment,
+  renderSeoGeoAuditWorkbook,
+} from "./seo-geo-audit-workbook";
+
+const csvPath = "docs/marketing/data/seo-geo-audit-workbook.csv";
+const requiredHeaders = [
+  "audit_id",
+  "baseline_id",
+  "experiment_id",
+  "route_role",
+  "canonical_path",
+  "target_user",
+  "query",
+  "query_cluster",
+  "surface",
+  "task_kind",
+  "evidence_fields",
+  "success_criteria",
+  "forbidden_claims",
+  "status",
+  "next_action",
+] as const;
+
+function parseRows() {
+  const source = readFileSync(join(process.cwd(), csvPath), "utf8").trim();
+  const [headerLine, ...lines] = source.split(/\r?\n/);
+  const headers = headerLine.split(",");
+  const rows = lines.map((line) => {
+    const values = line.split(",");
+    return Object.fromEntries(headers.map((header, index) => [header, values[index] ?? ""]));
+  });
+
+  return { headers, rows };
+}
+
+describe("SEO/GEO audit workbook", () => {
+  it("expands every Search/AI baseline row into one audit task per surface", () => {
+    const rows = buildSeoGeoAuditWorkbookRows();
+    const expectedCount = searchAiBaselineRows.reduce((count, row) => count + row.surfaces.length, 0);
+
+    expect(rows).toHaveLength(expectedCount);
+
+    for (const baseline of searchAiBaselineRows) {
+      for (const surface of baseline.surfaces) {
+        const row = rows.find((candidate) => candidate.auditId === `${baseline.id}_${surface}`);
+
+        expect(row).toEqual(
+          expect.objectContaining({
+            baselineId: baseline.id,
+            experimentId: baseline.experimentId,
+            routeRole: baseline.routeRole,
+            canonicalPath: baseline.canonicalPath,
+            targetUser: baseline.targetUser,
+            query: baseline.query,
+            queryCluster: baseline.queryCluster,
+            surface,
+            expectedTerms: baseline.expectedTerms,
+            forbiddenClaims: baseline.forbiddenClaims,
+            status: "pending",
+          }),
+        );
+        expect(row?.evidenceFields.length).toBeGreaterThan(4);
+        expect(row?.successCriteria).not.toContain("validated");
+        expect(row?.nextAction).not.toHaveLength(0);
+      }
+    }
+  });
+
+  it("uses Search Console and AI answer task kinds with surface-specific evidence fields", () => {
+    const rows = buildSeoGeoAuditWorkbookRows();
+    const searchConsoleRows = rows.filter((row) => row.surface === "google_search_console");
+    const aiRows = rows.filter((row) => row.taskKind === "ai_answer_audit");
+
+    expect(searchConsoleRows).toHaveLength(searchAiBaselineRows.length);
+    expect(aiRows).toHaveLength(searchAiBaselineRows.length * aiSearchAuditSurfaces.length);
+
+    for (const row of searchConsoleRows) {
+      expect(row.taskKind).toBe("search_console_query_baseline");
+      expect(row.evidenceFields).toEqual(
+        expect.arrayContaining([
+          "window_start",
+          "window_end",
+          "query",
+          "page",
+          "impressions",
+          "clicks",
+          "ctr",
+          "average_position",
+        ]),
+      );
+      expect(row.successCriteria).toBe("query_data_collected_and_mapped_to_canonical_page");
+    }
+
+    for (const surface of aiSearchAuditSurfaces) {
+      const surfaceRows = rows.filter((row) => row.surface === surface);
+
+      expect(surfaceRows).toHaveLength(searchAiBaselineRows.length);
+      for (const row of surfaceRows) {
+        expect(row.taskKind).toBe("ai_answer_audit");
+        expect(row.evidenceFields).toEqual(
+          expect.arrayContaining([
+            "audit_date",
+            "mentioned",
+            "cited",
+            "source_url",
+            "accuracy_status",
+            "competitors_cited",
+            "forbidden_claims_observed",
+          ]),
+        );
+        expect(row.successCriteria).toBe("answer_mentions_or_omits_quickfork_without_forbidden_claims");
+      }
+    }
+  });
+
+  it("mirrors the editable CSV workbook", () => {
+    const { headers, rows } = parseRows();
+    const workbookRows = buildSeoGeoAuditWorkbookRows();
+
+    expect(headers).toEqual([...requiredHeaders]);
+    expect(rows).toHaveLength(workbookRows.length);
+
+    for (const row of rows) {
+      const workbook = workbookRows.find((candidate) => candidate.auditId === row.audit_id);
+
+      expect(workbook).toEqual(
+        expect.objectContaining({
+          baselineId: row.baseline_id,
+          experimentId: row.experiment_id,
+          routeRole: row.route_role,
+          canonicalPath: row.canonical_path,
+          targetUser: row.target_user,
+          query: row.query,
+          queryCluster: row.query_cluster,
+          surface: row.surface as SearchBaselineSurface,
+          taskKind: row.task_kind,
+          status: row.status,
+          nextAction: row.next_action,
+        }),
+      );
+      expect(workbook?.evidenceFields.join("|")).toBe(row.evidence_fields);
+      expect(workbook?.successCriteria).toBe(row.success_criteria);
+      expect(workbook?.forbiddenClaims.join("|")).toBe(row.forbidden_claims);
+    }
+  });
+
+  it("renders an experiment workbook for supervisor audit without claiming validation", () => {
+    const runbook = renderSeoGeoAuditWorkbook("2026_q2_launch_materials_map_intent_validation");
+
+    expect(runbook).toContain("# SEO/GEO Evidence Workbook");
+    expect(runbook).toContain("Experiment: 2026_q2_launch_materials_map_intent_validation");
+    expect(runbook).toContain("/product/cold-start-launch-materials");
+    expect(runbook).toContain("/product/github-repo-launch-materials-map");
+    expect(runbook).toContain("google_search_console");
+    expect(runbook).toContain("chatgpt_search");
+    expect(runbook).toContain("perplexity");
+    expect(runbook).toContain("google_ai_overview");
+    expect(runbook).toContain("gemini");
+    expect(runbook).toContain("claude");
+    expect(runbook).toContain("impressions");
+    expect(runbook).toContain("mentioned");
+    expect(runbook).toContain("forbidden_claims_observed");
+    expect(runbook).toContain(
+      "Decision: insufficient_data_until_14_day_ga4_search_console_and_ai_answer_evidence_exists",
+    );
+    expect(runbook).not.toMatch(/validated demand|ranking win|revenue lift|customer traction|guaranteed launch/i);
+  });
+
+  it("keeps experiment workbooks scoped to registered baseline surfaces", () => {
+    const rows = getSeoGeoAuditWorkbookRowsForExperiment("2026_q2_source_backed_assets_intent_validation");
+
+    expect(rows).toHaveLength(2 * searchBaselineSurfaces.length);
+    expect(new Set(rows.map((row) => row.surface))).toEqual(new Set(searchBaselineSurfaces));
+    expect(rows.every((row) => row.experimentId === "2026_q2_source_backed_assets_intent_validation")).toBe(true);
+    expect(JSON.stringify(rows)).not.toMatch(/email|token|secret|password|api_key|revenue_amount|customer_logo/i);
+  });
+});
diff --git a/src/marketing/seo-geo-audit-workbook.ts b/src/marketing/seo-geo-audit-workbook.ts
new file mode 100644
index 0000000..d85e847
--- /dev/null
+++ b/src/marketing/seo-geo-audit-workbook.ts
@@ -0,0 +1,128 @@
+import { getGrowthExperimentById } from "./growth-experiments";
+import {
+  searchAiBaselineRows,
+  type ForbiddenAiSearchClaim,
+  type SearchAiBaselineRow,
+  type SearchBaselineSurface,
+} from "./search-ai-baseline";
+
+export const seoGeoAuditStatuses = ["pending", "collected", "needs_reaudit"] as const;
+export const seoGeoAuditTaskKinds = ["search_console_query_baseline", "ai_answer_audit"] as const;
+
+export type SeoGeoAuditStatus = (typeof seoGeoAuditStatuses)[number];
+export type SeoGeoAuditTaskKind = (typeof seoGeoAuditTaskKinds)[number];
+
+export interface SeoGeoAuditWorkbookRow {
+  auditId: string;
+  baselineId: string;
+  experimentId: string;
+  routeRole: SearchAiBaselineRow["routeRole"];
+  canonicalPath: string;
+  targetUser: string;
+  query: string;
+  queryCluster: string;
+  surface: SearchBaselineSurface;
+  taskKind: SeoGeoAuditTaskKind;
+  evidenceFields: readonly string[];
+  successCriteria: string;
+  expectedTerms: readonly string[];
+  forbiddenClaims: readonly ForbiddenAiSearchClaim[];
+  status: SeoGeoAuditStatus;
+  nextAction: string;
+}
+
+const searchConsoleEvidenceFields = [
+  "window_start",
+  "window_end",
+  "query",
+  "page",
+  "country",
+  "device",
+  "impressions",
+  "clicks",
+  "ctr",
+  "average_position",
+] as const;
+
+const aiAnswerEvidenceFields = [
+  "audit_date",
+  "mentioned",
+  "cited",
+  "source_url",
+  "answer_summary",
+  "accuracy_status",
+  "competitors_cited",
+  "forbidden_claims_observed",
+] as const;
+
+export function buildSeoGeoAuditWorkbookRows() {
+  return searchAiBaselineRows.flatMap((baseline) =>
+    baseline.surfaces.map((surface): SeoGeoAuditWorkbookRow => {
+      const isSearchConsole = surface === "google_search_console";
+
+      return {
+        auditId: `${baseline.id}_${surface}`,
+        baselineId: baseline.id,
+        experimentId: baseline.experimentId,
+        routeRole: baseline.routeRole,
+        canonicalPath: baseline.canonicalPath,
+        targetUser: baseline.targetUser,
+        query: baseline.query,
+        queryCluster: baseline.queryCluster,
+        surface,
+        taskKind: isSearchConsole ? "search_console_query_baseline" : "ai_answer_audit",
+        evidenceFields: isSearchConsole ? searchConsoleEvidenceFields : aiAnswerEvidenceFields,
+        successCriteria: isSearchConsole
+          ? "query_data_collected_and_mapped_to_canonical_page"
+          : "answer_mentions_or_omits_quickfork_without_forbidden_claims",
+        expectedTerms: baseline.expectedTerms,
+        forbiddenClaims: baseline.forbiddenClaims,
+        status: "pending",
+        nextAction: isSearchConsole
+          ? "export_14_day_search_console_query_page_metrics"
+          : "run_prompt_and_record_mention_citation_accuracy_and_competitors",
+      };
+    }),
+  );
+}
+
+export function getSeoGeoAuditWorkbookRowsForExperiment(experimentId: string) {
+  return buildSeoGeoAuditWorkbookRows().filter((row) => row.experimentId === experimentId);
+}
+
+export function renderSeoGeoAuditWorkbook(experimentId: string) {
+  const experiment = getGrowthExperimentById(experimentId);
+  const rows = getSeoGeoAuditWorkbookRowsForExperiment(experimentId);
+
+  if (!experiment || rows.length === 0) {
+    throw new Error(`Unknown SEO/GEO audit workbook experiment: ${experimentId}`);
+  }
+
+  return [
+    "# SEO/GEO Evidence Workbook",
+    "",
+    `Experiment: ${experiment.id}`,
+    `Lifecycle stage: ${experiment.lifecycleStage}`,
+    `Target user: ${experiment.targetUser}`,
+    `Minimum window: ${experiment.minimumWindow}`,
+    `Primary CTA: ${experiment.primaryCta}`,
+    `Primary metric: ${experiment.primaryMetric}`,
+    `Guardrail metric: ${experiment.guardrailMetric}`,
+    "Decision: insufficient_data_until_14_day_ga4_search_console_and_ai_answer_evidence_exists",
+    "",
+    "| Role | Route | Surface | Query | Task kind | Evidence fields | Success criteria | Next action |",
+    "| --- | --- | --- | --- | --- | --- | --- | --- |",
+    ...rows.map(
+      (row) =>
+        `| ${row.routeRole} | ${row.canonicalPath} | ${row.surface} | ${row.query} | ${row.taskKind} | ${row.evidenceFields.join(
+          "; ",
+        )} | ${row.successCriteria} | ${row.nextAction} |`,
+    ),
+    "",
+    "Guardrails:",
+    "",
+    "- Keep GA4, Search Console, and AI-answer evidence separate from the decision until the minimum window is complete.",
+    "- Record omissions as evidence; do not convert prompt coverage into visibility claims.",
+    `- Forbidden claims to flag: ${Array.from(new Set(rows.flatMap((row) => row.forbiddenClaims))).join("; ")}.`,
+  ].join("\n");
+}
diff --git a/src/server/generation/generation.test.ts b/src/server/generation/generation.test.ts
index 29bd247..6d55a37 100644
--- a/src/server/generation/generation.test.ts
+++ b/src/server/generation/generation.test.ts
@@ -547,6 +547,7 @@ describe("project launch generation backend", () => {
         "readme",
         "social",
         "deck",
+        "site",
         "visual",
         "outreach",
       ]);
@@ -561,7 +562,7 @@ describe("project launch generation backend", () => {
           source: expect.stringContaining("README or repo metadata includes"),
         }),
       );
-      expect(result.launchBrief.launchMaterialsMap.channels[3]).toEqual(
+      expect(result.launchBrief.launchMaterialsMap.channels[4]).toEqual(
         expect.objectContaining({
           type: "visual",
           artifactLabel: "Visual explainer prompt",
@@ -575,6 +576,7 @@ describe("project launch generation backend", () => {
         "readme",
         "social",
         "deck",
+        "site",
         "outreach",
         "visual",
       ]);
@@ -613,6 +615,13 @@ describe("project launch generation backend", () => {
             label: "Pitch deck outline",
             body: expect.stringContaining("Problem:"),
           }),
+          expect.objectContaining({
+            type: "site",
+            label: "Launch site page spine",
+            fileName: "nexu-io-open-design-launch-site-page-spine.md",
+            body: expect.stringContaining("## Launch site page spine"),
+            sourceReferences: expect.arrayContaining([expect.stringContaining("README or repo metadata includes")]),
+          }),
         ]),
       );
       expect(JSON.stringify(result.launchBrief)).not.toMatch(/best|guaranteed|customers|revenue/i);
diff --git a/src/server/generation/launch-brief.ts b/src/server/generation/launch-brief.ts
index 220b9e7..e22deba 100644
--- a/src/server/generation/launch-brief.ts
+++ b/src/server/generation/launch-brief.ts
@@ -25,7 +25,7 @@ export function buildRepoLaunchBrief(input: {
   const insights = padItems(input.brief.keyInsights, [
     "Repository evidence is compressed into a clear launch story.",
     "Source-backed assets help reviewers separate facts from hypotheses.",
-    "The launch package can be reused across README, social, deck, and outreach surfaces.",
+    "The launch package can be reused across README, social, deck, site, and outreach surfaces.",
   ]).slice(0, 3);
   const workflow = padItems(input.brief.workflowSteps, ["Brief", "Generate", "Review"]).slice(0, 3);
   const audience = audienceHypothesis(input.metadata);
@@ -191,6 +191,29 @@ function buildLaunchBriefArtifacts(input: {
       ].join("\n"),
       sourceReferences: input.sourceReferences,
     },
+    {
+      type: "site",
+      label: "Launch site page spine",
+      fileName: `${slug}-launch-site-page-spine.md`,
+      body: [
+        `# Launch site page spine for ${input.repoFullName}`,
+        "",
+        "## Launch site page spine",
+        "",
+        `Summary: ${input.summary}`,
+        `Audience hypothesis: ${input.audienceHypothesis}`,
+        "",
+        "## Page outline",
+        `- Hero: Explain ${input.repoFullName} with the strongest source-backed launch angle.`,
+        `- Proof: Use "${input.launchAngles[0]?.body ?? input.summary}" only if the cited source supports it.`,
+        "- Workflow: Show how a visitor moves from repository context to README, social, deck, visual, and outreach artifacts.",
+        "- CTA: Ask visitors to generate a similar launch package or request a reviewed full package.",
+        "- Guardrail: Do not publish unsupported traction, adoption, commercial, pricing, or benchmark claims.",
+        "",
+        sourceBlock,
+      ].join("\n"),
+      sourceReferences: input.sourceReferences,
+    },
     {
       type: "outreach",
       label: "Product outreach draft",
@@ -284,7 +307,7 @@ function buildLaunchMaterialsMap(input: {
 
   return {
     title: `${input.repoFullName} launch materials map`,
-    summary: `Channel plan for README, social, deck, visual, and outreach launch materials from ${input.summary}`,
+    summary: `Channel plan for README, social, deck, site, visual, and outreach launch materials from ${input.summary}`,
     channels: [
       {
         type: "readme",
@@ -319,6 +342,17 @@ function buildLaunchMaterialsMap(input: {
         reviewQuestion: "Which deck slide needs the strongest source citation before sharing outside the team?",
         successSignal: "Deck artifact copied or downloaded; full launch package CTA clicked after artifact review.",
       },
+      {
+        type: "site",
+        label: "Launch site page spine",
+        primaryUser: "Technical founders, DevRel operators, and visitors comparing the project before opening GitHub",
+        jobToBeDone: "Understand the project story, source proof, and launch package path before diving into the repository.",
+        artifactLabel: "Launch site page spine",
+        channelFit: `Turn the story map into a crawlable page outline: hero proof from "${topAngle}", workflow, artifacts, and review guardrails.`,
+        source: primarySource,
+        reviewQuestion: "Which source-backed proof belongs above the fold before publishing a public site page?",
+        successSignal: "Site spine copied or downloaded; full launch package request or generation_started follows the site CTA.",
+      },
       {
         type: "visual",
         label: "Visual project explainer",
@@ -394,7 +428,7 @@ function buildStoryMap(input: {
         id: "launch",
         label: "Launch",
         title: "Channel package",
-        detail: `Package README, social, deck, outreach, and visual assets using the ${input.visualCategory} direction.`,
+        detail: `Package README, social, deck, site, outreach, and visual assets using the ${input.visualCategory} direction.`,
         source: "Launch package channels from the QuickFork generation workflow.",
       },
     ],
diff --git a/src/server/generation/types.ts b/src/server/generation/types.ts
index 1c6eefb..834c7de 100644
--- a/src/server/generation/types.ts
+++ b/src/server/generation/types.ts
@@ -228,7 +228,7 @@ export interface RepoLaunchStoryMap {
   nodes: RepoLaunchStoryMapNode[];
 }
 
-export type RepoLaunchMaterialChannelType = "readme" | "social" | "deck" | "visual" | "outreach";
+export type RepoLaunchMaterialChannelType = "readme" | "social" | "deck" | "site" | "visual" | "outreach";
 
 export interface RepoLaunchMaterialChannel {
   type: RepoLaunchMaterialChannelType;
@@ -248,7 +248,7 @@ export interface RepoLaunchMaterialsMap {
   channels: RepoLaunchMaterialChannel[];
 }
 
-export type RepoLaunchBriefArtifactType = "audience" | "story_map" | "materials_map" | "readme" | "social" | "deck" | "outreach" | "visual";
+export type RepoLaunchBriefArtifactType = "audience" | "story_map" | "materials_map" | "readme" | "social" | "deck" | "site" | "outreach" | "visual";
 
 export interface RepoLaunchBriefArtifact {
   type: RepoLaunchBriefArtifactType;
diff --git a/src/server/marketing/launch-package-fit.test.ts b/src/server/marketing/launch-package-fit.test.ts
index 838475e..bff21fa 100644
--- a/src/server/marketing/launch-package-fit.test.ts
+++ b/src/server/marketing/launch-package-fit.test.ts
@@ -11,7 +11,7 @@ describe("launch package fit scoring", () => {
         launchTimeline: "within_30_days",
         packageModel: "single_launch",
         buyingTrigger: "launch_deadline",
-        packageScope: ["readme", "social", "deck", "outreach", "visual_explainer"],
+        packageScope: ["readme", "social", "deck", "outreach", "sites", "visual_explainer"],
         humanReviewNeeded: true,
         notes: "Launching an AI repo and need source-backed README, deck, and outreach review.",
       }),
diff --git a/src/server/marketing/lead-capture.test.ts b/src/server/marketing/lead-capture.test.ts
index 28dbbaa..9d3b36c 100644
--- a/src/server/marketing/lead-capture.test.ts
+++ b/src/server/marketing/lead-capture.test.ts
@@ -114,7 +114,7 @@ describe("lead capture", () => {
           launchTimeline: "within_30_days",
           packageModel: "single_launch",
           buyingTrigger: "launch_deadline",
-          packageScope: ["readme", "social", "deck", "outreach", "visual_explainer"],
+          packageScope: ["readme", "social", "deck", "outreach", "sites", "visual_explainer"],
           humanReviewNeeded: true,
           notes: "Launching an AI repo and need source-backed README, deck, and outreach review.",
         },
@@ -159,7 +159,7 @@ describe("lead capture", () => {
             launchTimeline: "within_30_days",
             packageModel: "single_launch",
             buyingTrigger: "launch_deadline",
-            packageScope: ["readme", "social", "deck", "outreach", "visual_explainer"],
+            packageScope: ["readme", "social", "deck", "outreach", "sites", "visual_explainer"],
             humanReviewNeeded: true,
             notes: "Launching an AI repo and need source-backed README, deck, and outreach review.",
           },