From 86668740edaf4b4bb7bc96b361d01db720e4983a Mon Sep 17 00:00:00 2001
From: mavneox <mav@onlywithneo.com>
Date: Fri, 24 Jul 2026 09:16:36 +0000
Subject: [PATCH 1/7] =?UTF-8?q?docs:=20sync=20API=20surface=20to=20142=20M?=
 =?UTF-8?q?CP=20tools=20=E2=80=94=20add=20billing,=20social,=20schedule,?=
 =?UTF-8?q?=20asset=20OpenAPI=20paths?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

- Update llms.txt + llms-full.txt: tool count 150→142 (precise audit)
- Add python.gen.pro to OpenAPI servers list
- Add Billing and Voices tags
- Add OpenAPI paths: credit_balance, credit_plans, credit_transactions, payment,
  subscription_upgrade, user_jobs/estimate, user_voice_resources,
  hume_ai_sample/status, content_resources/from_url, schedule/posts,
  schedule/with-post, schedule/post/{id} (GET+PATCH+DELETE),
  platform/get_oauth_account, platform/get_authorization_url,
  platform/disconnect/{platform}, agents/{id}/recurring-jobs/{id}/test-run

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
---
 public/llms-full.txt |   2 +-
 public/llms.txt      |   2 +-
 public/openapi.yaml  | 463 +++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 465 insertions(+), 2 deletions(-)

diff --git a/public/llms-full.txt b/public/llms-full.txt
index 3d2cd8a..1afba8c 100644
--- a/public/llms-full.txt
+++ b/public/llms-full.txt
@@ -1332,7 +1332,7 @@ next cadence tick.
 
 # Integrations
 
-- MCP Server: remote endpoint `https://mcp.gen.pro/mcp` with `Authorization: Bearer <pat>`, or local stdio package `uvx gen-mcp-server` (Python, PyPI; env `GEN_API_KEY`). Exposes 100 gen_* tools, 9 prompts, and 3 guidance resources.
+- MCP Server: remote endpoint `https://mcp.gen.pro/mcp` with `Authorization: Bearer <pat>`, or local stdio package `uvx gen-mcp-server` (Python, PyPI; env `GEN_API_KEY`). Exposes 142 gen_* tools, 9 prompts, and 3 guidance resources.
 - TypeScript SDK: `npm install @poweredbygen/gen-sdk`
 - n8n: HTTP Request node with `X-API-Key` header
 - OpenAPI 3.1: https://api.gen.pro/openapi.yaml  (every operation carries `x-phase`)
diff --git a/public/llms.txt b/public/llms.txt
index 334e6d1..ab4c44f 100644
--- a/public/llms.txt
+++ b/public/llms.txt
@@ -795,7 +795,7 @@ Local install example:
 GEN_API_KEY=your-pat uvx gen-mcp-server
 ```
 
-The current MCP exposes 150 `gen_*` tools, 9 prompts, and 3 guidance resources. Key tools by category:
+The current MCP exposes 142 `gen_*` tools, 9 prompts, and 3 guidance resources. Key tools by category:
 
 **Natural-language delegation (start here for fuzzy goals):** `gen_ask` (questions about social data — links, accounts, hashtags, trends), `gen_chat` (open-ended build goals), `gen_research` (topic research across social + web), `gen_make_video` (finished multi-scene video from a brief), `gen_generate_content_ideas`.
 
diff --git a/public/openapi.yaml b/public/openapi.yaml
index ac4757f..41126a1 100644
--- a/public/openapi.yaml
+++ b/public/openapi.yaml
@@ -59,6 +59,8 @@ servers:
     description: Agent Chat API
   - url: https://agent-core.gen.pro/v1
     description: Agent Core API (watchlists)
+  - url: https://python.gen.pro
+    description: Social publishing + OAuth integration service (schedule, social connect)
 
 security:
   - apiKey: []
@@ -125,6 +127,13 @@ tags:
       agent. Base URL `https://agent-core.gen.pro/v1`. Scraped data feeds back
       into agent chat and content idea generation. CRUD operations are free;
       downstream scrapes bill normally.
+  - name: Billing
+    description: |
+      Credit balance, usage history, plan discovery, and purchase flows. All on `https://api.gen.pro/v1`.
+      All endpoints that consume credits are marked `x-phase: export` and documented with a credit note.
+  - name: Voices
+    description: |
+      Voice library, custom voice CRUD, and voice-design status polling. Base URL `https://api.gen.pro/v1`.
   - name: Recurring Jobs
     description: |
       Scheduled agent runs ("daily tasks"). Base URL
@@ -4152,6 +4161,460 @@ paths:
             application/json:
               schema: { $ref: '#/components/schemas/RecurringJob' }
 
+  /agents/{agent_id}/recurring-jobs/{job_id}/test-run:
+    post:
+      operationId: runRecurringJobNow
+      x-phase: monitoring
+      summary: Trigger a recurring job immediately (test run)
+      tags: [Recurring Jobs]
+      servers:
+        - url: https://agent.gen.pro/v1
+      parameters:
+        - { name: agent_id, in: path, required: true, schema: { type: string } }
+        - { name: job_id, in: path, required: true, schema: { type: string } }
+      responses:
+        '202':
+          description: Run accepted
+          content:
+            application/json:
+              schema:
+                type: object
+                properties:
+                  run_id: { type: string }
+
+  /credit_balance:
+    get:
+      operationId: getCreditBalance
+      x-phase: export
+      summary: Get workspace credit balance
+      tags: [Billing]
+      parameters:
+        - { name: agent_id, in: query, required: true, schema: { type: string } }
+      responses:
+        '200':
+          description: Credit balance
+          content:
+            application/json:
+              schema:
+                type: object
+                properties:
+                  available_credit: { type: number }
+                  generic: { type: number }
+                  aura: { type: number }
+
+  /credit_plans:
+    get:
+      operationId: listCreditPlans
+      x-phase: export
+      summary: List available credit and subscription plans
+      tags: [Billing]
+      parameters:
+        - { name: cycle, in: query, schema: { type: string, enum: [monthly, annual] } }
+      responses:
+        '200':
+          description: Credit plans
+          content:
+            application/json:
+              schema:
+                type: array
+                items:
+                  type: object
+                  properties:
+                    id: { type: string }
+                    name: { type: string }
+                    price: { type: number }
+                    credits: { type: number }
+                    cycle: { type: string }
+
+  /credit_transactions:
+    get:
+      operationId: getCreditUsage
+      x-phase: export
+      summary: List credit usage history
+      tags: [Billing]
+      parameters:
+        - { name: agent_id, in: query, required: true, schema: { type: string } }
+        - { name: page, in: query, schema: { type: integer } }
+      responses:
+        '200':
+          description: Credit transactions
+          content:
+            application/json:
+              schema:
+                type: array
+                items:
+                  type: object
+                  properties:
+                    id: { type: string }
+                    amount: { type: number }
+                    description: { type: string }
+                    created_at: { type: string, format: date-time }
+
+  /payment:
+    post:
+      operationId: buyCredits
+      x-phase: export
+      summary: Start a Stripe credit purchase checkout session
+      tags: [Billing]
+      requestBody:
+        required: true
+        content:
+          application/json:
+            schema:
+              type: object
+              required: [credit_plan_id, organization_id]
+              properties:
+                credit_plan_id: { type: string }
+                organization_id: { type: string }
+                success_url: { type: string }
+                cancel_url: { type: string }
+      responses:
+        '200':
+          description: Checkout session with payment URL
+          content:
+            application/json:
+              schema:
+                type: object
+                properties:
+                  url: { type: string }
+
+  /subscription_upgrade:
+    post:
+      operationId: upgradeWorkspace
+      x-phase: export
+      summary: Upgrade workspace subscription tier
+      tags: [Billing]
+      requestBody:
+        required: true
+        content:
+          application/json:
+            schema:
+              type: object
+              required: [creditPlanId, organizationId, returnUrl]
+              properties:
+                creditPlanId: { type: string }
+                organizationId: { type: string }
+                returnUrl: { type: string }
+      responses:
+        '200':
+          description: Checkout session
+          content:
+            application/json:
+              schema:
+                type: object
+                properties:
+                  url: { type: string }
+
+  /user_jobs/estimate:
+    post:
+      operationId: estimateJob
+      x-phase: edit
+      summary: Estimate credit cost for one or more generation jobs
+      tags: [Billing]
+      requestBody:
+        required: true
+        content:
+          application/json:
+            schema:
+              type: object
+              required: [agent_id, user_jobs]
+              properties:
+                agent_id: { type: string }
+                user_jobs:
+                  type: array
+                  items: { type: object }
+      responses:
+        '200':
+          description: Cost estimate
+          content:
+            application/json:
+              schema:
+                type: object
+                properties:
+                  total_credits: { type: number }
+                  breakdown: { type: array, items: { type: object } }
+
+  /user_voice_resources:
+    get:
+      operationId: listMyVoices
+      x-phase: setup
+      summary: List the agent's custom voices
+      tags: [Voices]
+      parameters:
+        - { name: agent_id, in: query, required: true, schema: { type: string } }
+      responses:
+        '200':
+          description: Custom voices
+          content:
+            application/json:
+              schema:
+                type: array
+                items:
+                  type: object
+                  properties:
+                    id: { type: string }
+                    name: { type: string }
+                    source: { type: string }
+
+  /hume_ai_sample/status:
+    get:
+      operationId: getVoiceSampleStatus
+      x-phase: setup
+      summary: Poll voice design sample status
+      tags: [Voices]
+      parameters:
+        - { name: agent_id, in: query, required: true, schema: { type: string } }
+        - { name: generation_id, in: query, required: true, schema: { type: string } }
+      responses:
+        '200':
+          description: Sample status and audio
+          content:
+            application/json:
+              schema:
+                type: object
+                properties:
+                  status: { type: string, enum: [pending, completed, failed] }
+                  base64_audio: { type: string }
+                  speaker_embedding_url: { type: string }
+
+  /content_resources/from_url:
+    post:
+      operationId: importAssetFromUrl
+      x-phase: edit
+      summary: Import media from a public URL into the asset library
+      tags: [Content Resources]
+      requestBody:
+        required: true
+        content:
+          application/json:
+            schema:
+              type: object
+              required: [agent_id, url]
+              properties:
+                agent_id: { type: string }
+                url: { type: string, description: "Public URL (YouTube, TikTok, Instagram, or direct file)" }
+                folder_id: { type: string }
+      responses:
+        '200':
+          description: Import job started
+          content:
+            application/json:
+              schema:
+                type: object
+                properties:
+                  id: { type: string }
+                  status: { type: string }
+
+  /schedule/posts:
+    get:
+      operationId: listScheduledPosts
+      x-phase: export
+      summary: List the content calendar (scheduled posts)
+      tags: [Publishing]
+      servers:
+        - url: https://python.gen.pro
+      parameters:
+        - { name: agent_id, in: query, required: true, schema: { type: string } }
+        - { name: start_date, in: query, schema: { type: string, format: date } }
+        - { name: end_date, in: query, schema: { type: string, format: date } }
+      responses:
+        '200':
+          description: Scheduled posts
+          content:
+            application/json:
+              schema:
+                type: object
+                properties:
+                  scheduled_posts:
+                    type: array
+                    items:
+                      type: object
+
+  /schedule/with-post:
+    post:
+      operationId: schedulePost
+      x-phase: export
+      summary: Create or schedule a social media post
+      tags: [Publishing]
+      servers:
+        - url: https://python.gen.pro
+      requestBody:
+        required: true
+        content:
+          application/json:
+            schema:
+              type: object
+              required: [agent_id, platform, schedule_type]
+              properties:
+                agent_id: { type: string }
+                platform:
+                  type: array
+                  items: { type: string, enum: [tiktok, instagram, facebook, youtube, x] }
+                description: { type: string }
+                media_url: { type: string }
+                media_urls: { type: array, items: { type: string } }
+                schedule_type: { type: string, enum: [now, specific_time] }
+                scheduled_time: { type: string, format: date-time }
+                title: { type: string }
+                media_type: { type: string, enum: [VIDEO, IMAGE, TWEET_VIDEO] }
+                thumbnail_url: { type: string }
+                thread:
+                  type: array
+                  items:
+                    type: object
+                    properties:
+                      text: { type: string }
+                      media_urls: { type: array, items: { type: string } }
+                reply_to_tweet_id: { type: string }
+      responses:
+        '200':
+          description: Scheduled post created
+          content:
+            application/json:
+              schema:
+                type: object
+                properties:
+                  id: { type: string }
+                  status: { type: string }
+
+  /schedule/post/{post_id}:
+    get:
+      operationId: getPostStatus
+      x-phase: export
+      summary: Get platform-confirmed publish status of a post
+      tags: [Publishing]
+      servers:
+        - url: https://python.gen.pro
+      parameters:
+        - { name: post_id, in: path, required: true, schema: { type: string } }
+      responses:
+        '200':
+          description: Post status
+          content:
+            application/json:
+              schema:
+                type: object
+                properties:
+                  status: { type: string, enum: [accepted, publishing, published, failed] }
+                  platforms:
+                    type: array
+                    items:
+                      type: object
+                      properties:
+                        platform: { type: string }
+                        status: { type: string }
+                        platform_post_id: { type: string }
+                        post_url: { type: string }
+                        published_at: { type: string, format: date-time }
+                        error: { type: string }
+    patch:
+      operationId: updateScheduledPost
+      x-phase: export
+      summary: Update a scheduled post before it publishes
+      tags: [Publishing]
+      servers:
+        - url: https://python.gen.pro
+      parameters:
+        - { name: post_id, in: path, required: true, schema: { type: string } }
+      requestBody:
+        required: true
+        content:
+          application/json:
+            schema: { type: object, additionalProperties: true }
+      responses:
+        '200':
+          description: Updated post
+    delete:
+      operationId: deleteScheduledPost
+      x-phase: export
+      summary: Delete a scheduled post from the calendar
+      tags: [Publishing]
+      servers:
+        - url: https://python.gen.pro
+      parameters:
+        - { name: post_id, in: path, required: true, schema: { type: string } }
+      responses:
+        '204':
+          description: Deleted
+
+  /platform/get_oauth_account:
+    post:
+      operationId: listConnectedSocials
+      x-phase: export
+      summary: List connected social accounts for publishing
+      tags: [Publishing]
+      servers:
+        - url: https://python.gen.pro
+      requestBody:
+        required: true
+        content:
+          application/json:
+            schema:
+              type: object
+              required: [agent_id]
+              properties:
+                agent_id: { type: string }
+      responses:
+        '200':
+          description: Connected platforms
+          content:
+            application/json:
+              schema:
+                type: object
+                additionalProperties: true
+
+  /platform/get_authorization_url:
+    post:
+      operationId: getSocialConnectUrl
+      x-phase: export
+      summary: Get OAuth URL to connect a social account
+      tags: [Publishing]
+      servers:
+        - url: https://python.gen.pro
+      parameters:
+        - { name: platform, in: query, required: true, schema: { type: string, enum: [tiktok, instagram, facebook, youtube, x] } }
+      requestBody:
+        required: true
+        content:
+          application/json:
+            schema:
+              type: object
+              required: [agent_id]
+              properties:
+                agent_id: { type: string }
+      responses:
+        '200':
+          description: OAuth URL
+          content:
+            application/json:
+              schema:
+                type: object
+                properties:
+                  url: { type: string }
+
+  /platform/disconnect/{platform}:
+    put:
+      operationId: disconnectSocial
+      x-phase: export
+      summary: Disconnect a connected social account
+      tags: [Publishing]
+      servers:
+        - url: https://python.gen.pro
+      parameters:
+        - { name: platform, in: path, required: true, schema: { type: string, enum: [tiktok, instagram, facebook, youtube, x] } }
+      requestBody:
+        required: true
+        content:
+          application/json:
+            schema:
+              type: object
+              required: [agent_id]
+              properties:
+                agent_id: { type: string }
+      responses:
+        '200':
+          description: Disconnected
+
 components:
   securitySchemes:
     apiKey:

From cff23f15b66f2d597ab3d26f46997284059b65e6 Mon Sep 17 00:00:00 2001
From: mavneox <mav@onlywithneo.com>
Date: Sat, 25 Jul 2026 09:19:35 +0000
Subject: [PATCH 2/7] docs(llms.txt): sync MCP tool list to 142 live tools

The MCP has grown from 100 to 142 tools since the last sync. This
update fixes the tools section of llms.txt:
- Remove stale names: gen_research, gen_make_video, gen_list_voices,
  gen_create_voice (removed from live server)
- Add new tools: gen_run_research, gen_refine_content_ideas,
  gen_list_content_ideas, gen_update_idea_status, gen_expand_idea,
  gen_set_content_preference, gen_decide_agent_run,
  gen_list_conversations, gen_get_conversation,
  gen_get_session_preferences, gen_update_session_preferences,
  gen_generate_content, gen_generate_layer, gen_continue_generation,
  gen_publish_content, gen_list_agent_voices, gen_preview_voice,
  gen_get_voice_preview_status, gen_generate_voice_description,
  gen_create_content_resource, gen_get_content_resource,
  gen_list_content_resources, gen_delete_content_resource,
  gen_update_column, gen_duplicate_row, gen_create_layer,
  gen_get_layer, gen_update_layer, gen_delete_layer,
  gen_pause_watchlist, gen_resume_watchlist, gen_get_watchlist,
  gen_update_watchlist, gen_delete_watchlist, gen_create_monitoring_job,
  gen_update_monitoring_job, gen_list_proof_of_genesis_backups,
  gen_create_proof_of_genesis_backup, gen_remove_proof_of_genesis_backup,
  gen_get_recurring_job, gen_ensure_default_recurring_job,
  gen_get_credit_usage, gen_list_credit_plans, gen_update_agent,
  gen_get_agent, gen_delete_agent, gen_get_agent_profile,
  gen_create_agent_profile, gen_update_agent_profile,
  gen_reset_agent_profile, gen_add_agent_account,
  gen_add_agent_inspiration, gen_connect_agent_elevenlabs,
  gen_update_organization, gen_delete_organization,
  gen_list_api_keys, gen_update_api_key, gen_revoke_api_key,
  gen_revoke_all_api_keys

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
---
 public/llms.txt | 34 +++++++++++++++++++++++-----------
 1 file changed, 23 insertions(+), 11 deletions(-)

diff --git a/public/llms.txt b/public/llms.txt
index ab4c44f..030e323 100644
--- a/public/llms.txt
+++ b/public/llms.txt
@@ -797,27 +797,39 @@ GEN_API_KEY=your-pat uvx gen-mcp-server
 
 The current MCP exposes 142 `gen_*` tools, 9 prompts, and 3 guidance resources. Key tools by category:
 
-**Natural-language delegation (start here for fuzzy goals):** `gen_ask` (questions about social data — links, accounts, hashtags, trends), `gen_chat` (open-ended build goals), `gen_research` (topic research across social + web), `gen_make_video` (finished multi-scene video from a brief), `gen_generate_content_ideas`.
+**Natural-language delegation (start here for fuzzy goals):** `gen_ask` (questions about social data — links, accounts, hashtags, trends), `gen_chat` (open-ended build goals), `gen_run_research` (deep topic research across 10+ platforms), `gen_generate_content_ideas` (data-driven ideas/scripts), `gen_refine_content_ideas` (iterate with feedback).
 
-**Creation:** `gen_create_video` (single clip; models: seedance-2.0, veo_3, kling_2_1, sora_2, grok, pika), `gen_create_image` (nano-banana-pro/nano-banana/nano-banana-2), `gen_create_song` (suno-v5-beta/suno-v4.5plus-beta), `gen_create_song_mix`, `gen_render_video`.
+**Content ideas lifecycle:** `gen_list_content_ideas` (see all ideas), `gen_expand_idea` (expand one idea to a full scene plan), `gen_update_idea_status` (approve/reject/archive), `gen_set_content_preference` (persistent rules for future generations), `gen_get_session_preferences` / `gen_update_session_preferences` (per-conversation creative rules), `gen_decide_agent_run` (approve/reject a pending agent action gate).
 
-**Assets:** `gen_list_assets`, `gen_get_asset`, `gen_import_asset_from_url`, `gen_upload_asset`, `gen_complete_asset_upload`, `gen_delete_asset`.
+**Conversations:** `gen_list_conversations`, `gen_get_conversation`.
 
-**Voices:** `gen_list_voices`, `gen_list_my_voices`, `gen_design_voice`, `gen_clone_voice`, `gen_create_voice`.
+**Creation:** `gen_create_video` (single clip; models: seedance-2.0, veo_3, kling_2_1, sora_2, grok, pika), `gen_create_image` (nano-banana-pro/nano-banana/nano-banana-2), `gen_create_song` (suno-v5-beta/suno-v4.5plus-beta), `gen_create_song_mix`, `gen_render_video`, `gen_generate_content` (trigger generation for a vidsheet cell), `gen_generate_layer` (trigger generation for a specific layer), `gen_continue_generation` (resume a stopped generation).
 
-**Social publishing:** `gen_get_social_connect_url` (OAuth connect flow), `gen_list_connected_socials`, `gen_schedule_post` (platforms: tiktok|instagram|facebook|youtube|x), `gen_list_scheduled_posts`, `gen_get_post_status`, `gen_update_scheduled_post`, `gen_delete_scheduled_post`, `gen_disconnect_social`. On **X**, `gen_schedule_post` also supports up to 4 images per tweet (`media_urls`), **threads** (`thread`: an ordered list of `{text, media_urls?}` segments), and **replies** (`reply_to_tweet_id`). `thread` and `reply_to_tweet_id` are X-only.
+**Assets / Content Resources:** `gen_list_assets` / `gen_list_content_resources`, `gen_get_asset` / `gen_get_content_resource`, `gen_import_asset_from_url`, `gen_upload_asset`, `gen_complete_asset_upload`, `gen_create_content_resource`, `gen_update_content_resource`, `gen_delete_asset` / `gen_delete_content_resource`.
 
-**Vidsheets:** `gen_list_templates`, `gen_clone_template`, `gen_create_engine`, `gen_get_engine`, `gen_clone_engine`, `gen_list_columns`, `gen_create_column`, `gen_list_rows`, `gen_create_row`, `gen_get_cell`, `gen_update_cell`, `gen_create_variable`, `gen_update_variable`, `gen_delete_variable`.
+**Voices:** `gen_list_agent_voices`, `gen_list_my_voices`, `gen_design_voice`, `gen_clone_voice`, `gen_preview_voice` (TTS audition), `gen_get_voice_preview_status`, `gen_generate_voice_description`, `gen_generate_voice_samples`, `gen_get_voice_sample_status`, `gen_delete_voice`.
 
-**Watchlists:** `gen_create_watchlist`, `gen_list_watchlists`, `gen_query_watchlist`, `gen_add_watchlist_source`, `gen_remove_watchlist_source`.
+**Social publishing:** `gen_get_social_connect_url` (OAuth connect flow), `gen_list_connected_socials`, `gen_schedule_post` (platforms: tiktok|instagram|facebook|youtube|x), `gen_list_scheduled_posts`, `gen_get_post_status`, `gen_update_scheduled_post`, `gen_delete_scheduled_post`, `gen_disconnect_social`, `gen_publish_content` (direct publish/schedule to TikTok). On **X**, `gen_schedule_post` also supports up to 4 images per tweet (`media_urls`), **threads** (`thread`: an ordered list of `{text, media_urls?}` segments), and **replies** (`reply_to_tweet_id`). `thread` and `reply_to_tweet_id` are X-only.
 
-**Recurring automation:** `gen_create_recurring_job`, `gen_list_recurring_jobs`, `gen_run_recurring_job_now`, `gen_pause_recurring_job`, `gen_resume_recurring_job`.
+**Vidsheets:** `gen_list_templates`, `gen_clone_template`, `gen_create_engine`, `gen_get_engine`, `gen_clone_engine`, `gen_list_columns`, `gen_create_column`, `gen_update_column`, `gen_delete_column`, `gen_list_rows`, `gen_create_row`, `gen_duplicate_row`, `gen_get_cell`, `gen_update_cell`, `gen_create_layer`, `gen_get_layer`, `gen_update_layer`, `gen_delete_layer`, `gen_list_variables`, `gen_create_variable`, `gen_update_variable`, `gen_delete_variable`.
 
-**Credits/billing:** `gen_get_credit_balance`, `gen_estimate_job`, `gen_buy_credits`, `gen_upgrade_workspace`.
+**Watchlists:** `gen_create_watchlist`, `gen_list_watchlists`, `gen_get_watchlist`, `gen_update_watchlist`, `gen_delete_watchlist`, `gen_query_watchlist`, `gen_add_watchlist_source`, `gen_remove_watchlist_source`, `gen_pause_watchlist`, `gen_resume_watchlist`.
 
-**Agent setup:** `gen_get_me`, `gen_list_agents`, `gen_get_agent_core`, `gen_update_agent_core`, `gen_set_agent_primary_photo`, `gen_add_agent_look`.
+**Content monitoring:** `gen_create_monitoring_job` (scrape TikTok/Instagram/YouTube into agent knowledge), `gen_update_monitoring_job`.
 
-**Job control:** `gen_get_generation` (poll), `gen_stop_generation` (refunds credits), `gen_get_run_status`.
+**Proof of Genesis backups:** `gen_list_proof_of_genesis_backups`, `gen_create_proof_of_genesis_backup`, `gen_remove_proof_of_genesis_backup`.
+
+**Recurring automation:** `gen_create_recurring_job`, `gen_list_recurring_jobs`, `gen_get_recurring_job`, `gen_update_recurring_job`, `gen_run_recurring_job_now`, `gen_pause_recurring_job`, `gen_resume_recurring_job`, `gen_delete_recurring_job`, `gen_ensure_default_recurring_job` (idempotent daily-ideas setup).
+
+**Credits/billing:** `gen_get_credit_balance`, `gen_estimate_job`, `gen_get_credit_usage`, `gen_list_credit_plans`, `gen_buy_credits`, `gen_upgrade_workspace`.
+
+**Agent setup:** `gen_get_me`, `gen_list_agents`, `gen_get_agent`, `gen_create_agent`, `gen_update_agent`, `gen_delete_agent`, `gen_get_agent_core`, `gen_update_agent_core`, `gen_get_agent_profile`, `gen_create_agent_profile`, `gen_update_agent_profile`, `gen_reset_agent_profile`, `gen_set_agent_primary_photo`, `gen_add_agent_look`, `gen_add_agent_account`, `gen_add_agent_inspiration`, `gen_connect_agent_elevenlabs`, `gen_list_agent_avatars`, `gen_create_agent_avatar`, `gen_delete_agent_avatar`.
+
+**Organizations:** `gen_list_organizations`, `gen_create_organization`, `gen_get_organization`, `gen_update_organization`, `gen_delete_organization`.
+
+**API keys:** `gen_list_api_keys`, `gen_create_api_key`, `gen_update_api_key`, `gen_revoke_api_key`, `gen_revoke_all_api_keys`.
+
+**Job control:** `gen_get_generation` (poll), `gen_stop_generation` (refunds credits), `gen_continue_generation` (resume), `gen_get_run_status`.
 
 Full install guide: https://api.gen.pro/guides/mcp/
 

From b4ed36dee7a958bef34394d8515e59a08b859901 Mon Sep 17 00:00:00 2001
From: mavneox <mav@onlywithneo.com>
Date: Mon, 27 Jul 2026 09:14:50 +0000
Subject: [PATCH 3/7] api-sync 2026-07-27: surface sync + security fixes
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Security fixes:
- Remove pricing table ($0.015/video, $0.002/comment) from content-monitoring.mdx and llms-full.txt
- Replace "S3" with "cloud storage"/"from your client" in content-resources.mdx

Surface sync (MCP → docs/OpenAPI/llms):
- OpenAPI: add PATCH /vidsheet/{sheet_id}/columns/{column_id} for gen_update_column
- OpenAPI: add POST /agents/{agent_id}/core/inspiration for gen_add_agent_inspiration
- columns.mdx: add Update column section with params, errors, and curl example
- llms-full.txt: add PATCH/DELETE to columns section; add Agent Profile section

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
---
 public/llms-full.txt                          | 26 ++++++-
 public/openapi.yaml                           | 72 +++++++++++++++++++
 src/content/docs/reference/columns.mdx        | 40 +++++++++++
 .../docs/reference/content-monitoring.mdx     | 11 +--
 .../docs/reference/content-resources.mdx      |  4 +-
 5 files changed, 141 insertions(+), 12 deletions(-)

diff --git a/public/llms-full.txt b/public/llms-full.txt
index 1afba8c..cd16a19 100644
--- a/public/llms-full.txt
+++ b/public/llms-full.txt
@@ -294,6 +294,26 @@ DELETE /v1/persisted_tokens/{id}/revoke
 DELETE /v1/persisted_tokens/revoke_all
 ```
 
+## Agent Profile (agentic service mirror, `https://agent.gen.pro/v1`)
+
+The agent profile is a lightweight mirror of agent setup as seen by the ideas/research service. Use `gen_get_agent_core` / `gen_update_agent_core` (Rails) for the authoritative read/write. These endpoints are useful when you need to check or reset how the agentic service sees the agent, or to set `content_idea_preferences` (persistent rules for future idea runs).
+
+```
+GET    /v1/agent/profile?agent_id={id}
+→ { identity: { name, description, persona }, voice: { api_keys, default_voice }, brand: { brand_name, description, goal, keywords, target_platforms, shortform, longform, linked_accounts, content_idea_preferences } }
+
+POST   /v1/agent/profile?agent_id={id}
+Body: { identity?, voice?, brand? }
+→ updated profile
+
+PUT    /v1/agent/profile?agent_id={id}
+Body: { identity?, voice?, brand? }      (replaces; same shape as POST)
+
+DELETE /v1/agent/profile?agent_id={id}  — resets brand section only (does NOT delete the agent)
+```
+
+`content_idea_preferences` (inside `brand`) holds persistent plain-text rules applied to every future idea generation, e.g. `"- always use statement hooks\n- never mention competitors"`.
+
 ---
 
 # Phase 2 — Ideas
@@ -480,9 +500,9 @@ Only pending/processing jobs can be updated.
 | Instagram | Yes | Yes | No |
 | YouTube | Yes | Yes | Yes |
 
-### Pricing
+### Credit usage
 
-Per video/post: $0.015. Per comment (when enabled): $0.002.
+Content monitoring uses credits per post and comment scraped. Check your balance with `GET /v1/credit_balance?agent_id={id}` before starting large jobs.
 
 ### Reading scraped data
 
@@ -563,6 +583,8 @@ Types: `text`, `image`, `video`, `audio`.
 ```
 GET    /v1/vidsheet/{id}/columns?agent_id={id}
 POST   /v1/vidsheet/{id}/columns?agent_id={id}               Body: { spreadsheet_column: { title, type, position? } }
+PATCH  /v1/vidsheet/{id}/columns/{col_id}?agent_id={id}     Body: { spreadsheet_column: { title?, type?, position? } }
+DELETE /v1/vidsheet/{id}/columns/{col_id}?agent_id={id}
 POST   /v1/vidsheet/{id}/columns/{col_id}/duplicate?agent_id={id}
 PATCH  /v1/vidsheet/{id}/columns/update_positions?agent_id={id}
 PUT    /v1/vidsheet/{id}/columns/mass_update?agent_id={id}
diff --git a/public/openapi.yaml b/public/openapi.yaml
index 41126a1..4c804cd 100644
--- a/public/openapi.yaml
+++ b/public/openapi.yaml
@@ -974,6 +974,49 @@ paths:
               schema:
                 $ref: '#/components/schemas/Column'
 
+  /vidsheet/{sheet_id}/columns/{column_id}:
+    patch:
+      operationId: updateColumn
+      x-phase: edit
+      summary: Update a column
+      description: Update a column's title, type, or position. Only ingredient-role columns can be modified.
+      tags: [Columns]
+      parameters:
+        - $ref: '#/components/parameters/AgentId'
+        - $ref: '#/components/parameters/SheetId'
+        - name: column_id
+          in: path
+          required: true
+          schema:
+            type: string
+      requestBody:
+        content:
+          application/json:
+            schema:
+              type: object
+              properties:
+                spreadsheet_column:
+                  type: object
+                  properties:
+                    title:
+                      type: string
+                    type:
+                      type: string
+                      enum: [text, image, video, audio]
+                    position:
+                      type: integer
+      responses:
+        '200':
+          description: Column updated
+          content:
+            application/json:
+              schema:
+                $ref: '#/components/schemas/Column'
+        '403':
+          description: Forbidden — cannot modify system-role columns
+        '422':
+          description: Validation error
+
   /vidsheet/{sheet_id}/columns/{column_id}/duplicate:
     post:
       operationId: duplicateColumn
@@ -3410,6 +3453,35 @@ paths:
             application/json:
               schema: { $ref: '#/components/schemas/Error' }
 
+  /agents/{agent_id}/core/inspiration:
+    post:
+      operationId: addAgentInspiration
+      x-phase: setup
+      summary: Add one inspiration source to the agent
+      description: Adds a single inspiration source URL to the agent's existing inspiration list. Does not replace the full list. Use PATCH /agents/{agent_id}/core with the full `inspiration` array to replace.
+      tags: [Agent Core]
+      parameters:
+        - { name: agent_id, in: path, required: true, schema: { type: string } }
+      requestBody:
+        required: true
+        content:
+          application/json:
+            schema:
+              type: object
+              required: [url]
+              properties:
+                url:
+                  type: string
+                  description: Full URL of the inspiration source
+                platform:
+                  type: string
+                  description: Platform hint (e.g. tiktok, instagram, youtube)
+      responses:
+        '200':
+          description: Inspiration source added
+        '422':
+          description: Validation error
+
   # ── Agent Voice ───────────────────────────────────────────
   /agents/{agent_id}/proof-of-genesis/backups:
     get:
diff --git a/src/content/docs/reference/columns.mdx b/src/content/docs/reference/columns.mdx
index 61ab981..e0c5875 100644
--- a/src/content/docs/reference/columns.mdx
+++ b/src/content/docs/reference/columns.mdx
@@ -123,6 +123,46 @@ curl -X POST "https://api.gen.pro/v1/vidsheet/101/columns?agent_id=42" \
 
 ---
 
+## Update column
+
+Update a column's title, type, or position. Only `ingredient`-role columns can be updated.
+
+```
+PATCH /v1/vidsheet/{id}/columns/{col_id}?agent_id={agent_id}
+```
+
+### Request body
+
+| Field | Type | Required | Description |
+|-------|------|----------|-------------|
+| `spreadsheet_column.title` | string | No | New column name. |
+| `spreadsheet_column.type` | string | No | New content type: `text`, `image`, `video`, or `audio`. |
+| `spreadsheet_column.position` | integer | No | New zero-indexed position. |
+
+Send only the fields you want to change.
+
+### Response (200)
+
+Returns the updated column.
+
+### Errors
+
+| Status | Code | Description |
+|--------|------|-------------|
+| `403` | `forbidden` | Column is a system role and cannot be modified. |
+| `422` | `validation_error` | Invalid type or missing required field. |
+
+### Example
+
+```bash
+curl -X PATCH "https://api.gen.pro/v1/vidsheet/101/columns/200?agent_id=42" \
+  -H "X-API-Key: your-api-key" \
+  -H "Content-Type: application/json" \
+  -d '{"spreadsheet_column": {"title": "Voiceover Script", "position": 0}}'
+```
+
+---
+
 ## Duplicate column
 
 Creates a copy of a column and all its cell values.
diff --git a/src/content/docs/reference/content-monitoring.mdx b/src/content/docs/reference/content-monitoring.mdx
index 06d75e0..430b589 100644
--- a/src/content/docs/reference/content-monitoring.mdx
+++ b/src/content/docs/reference/content-monitoring.mdx
@@ -30,14 +30,9 @@ Not all platforms support all search types:
 | Instagram | Yes | Yes | No |
 | YouTube | Yes | Yes | Yes |
 
-## Pricing
-
-| Item | Cost |
-|------|------|
-| Per video/post scraped | $0.015 |
-| Per comment (when comment monitoring is enabled) | $0.002 |
-
-Costs are converted to credits using the standard credit formula.
+<Aside type="tip" title="Credit usage">
+  Content monitoring uses credits. One-time scrapes and ongoing monitoring both consume credits per post/comment scraped. Check your balance with `gen_get_credit_balance` before starting large monitoring jobs.
+</Aside>
 
 ---
 
diff --git a/src/content/docs/reference/content-resources.mdx b/src/content/docs/reference/content-resources.mdx
index d6cf794..5e7ee56 100644
--- a/src/content/docs/reference/content-resources.mdx
+++ b/src/content/docs/reference/content-resources.mdx
@@ -5,7 +5,7 @@ description: Upload, organize, and manage files — images, videos, audio, and f
 
 import { Aside } from '@astrojs/starlight/components';
 
-Content resources represent files in your agent's workspace — generated outputs (images, videos, audio) and manually uploaded assets. The asset library provides a unified view with folder organization, and direct upload lets you send files straight to S3 for maximum performance.
+Content resources represent files in your agent's workspace — generated outputs (images, videos, audio) and manually uploaded assets. The asset library provides a unified view with folder organization, and direct upload lets you send files straight from your client for maximum performance.
 
 <Aside type="note">
   All endpoints in this section require an `agent_id` query parameter to scope operations to a specific agent. See the [Discovery](/reference/discovery) page to list your agents.
@@ -465,7 +465,7 @@ curl -X DELETE "https://api.gen.pro/v1/asset_folders/16?agent_id=42" \
 
 ## Create direct upload
 
-Returns a pre-signed S3 URL for uploading a file directly from the client. This is the recommended approach for large files — the upload goes straight to S3 without passing through the API server.
+Returns a pre-signed URL for uploading a file directly from the client. This is the recommended approach for large files — the upload goes straight to cloud storage without passing through the API server.
 
 **Workflow:**
 1. Call this endpoint with file metadata to get a signed upload URL.

From 96fb7125701382df06a467de3f4de5a48afbd507 Mon Sep 17 00:00:00 2001
From: mavneox <mav@onlywithneo.com>
Date: Tue, 28 Jul 2026 09:25:46 +0000
Subject: [PATCH 4/7] docs: sync social publishing, new video models, and
 run-now endpoint

- Rewrite publishing.mdx: full social connect/disconnect OAuth flow,
  modern schedule/with-post (all 5 platforms), content calendar CRUD.
  Replaces outdated user_jobs publish_content TikTok-only pattern.
- Add "Run a job immediately" section to recurring-jobs.mdx with
  POST /v1/agents/:agent_id/recurring-jobs/:job_id/test-run
- Add Seedance 2.0 model row to video-from-text.mdx and video-from-image.mdx
- Update llms.txt Step 5 publishing section: python.gen.pro endpoints,
  all 5 platforms, OAuth connect/disconnect flow
- Update llms-full.txt Phase 5: replace old user_jobs pattern with
  modern social connect API + full schedule/with-post documentation;
  add test-run to recurring jobs section

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
---
 public/llms-full.txt                          | 143 ++++----
 public/llms.txt                               |  79 +++--
 .../docs/reference/cards/video-from-image.mdx |   1 +
 .../docs/reference/cards/video-from-text.mdx  |   3 +-
 src/content/docs/reference/publishing.mdx     | 307 ++++++++----------
 src/content/docs/reference/recurring-jobs.mdx |  17 +
 6 files changed, 293 insertions(+), 257 deletions(-)

diff --git a/public/llms-full.txt b/public/llms-full.txt
index cd16a19..d42e497 100644
--- a/public/llms-full.txt
+++ b/public/llms-full.txt
@@ -980,66 +980,88 @@ curl -L -o final_video.mp4 "https://cdn.gen.pro/outputs/render_xyz.mp4"
 
 CDN URLs are public and permanent until you explicitly delete the content resource. Individual layer outputs are also downloadable via each layer's `default_user_job.output_resources[0].url`.
 
+## Connect Social Accounts
+
+Base URL for social connect and scheduling: `https://python.gen.pro`
+
+Connect accounts with OAuth — never pass credentials through the API.
+
+```
+GET  /platform/get_authorization_url?agent_id={id}&platform={platform}
+→ { url: "https://..." }   ← user opens this in a browser
+```
+
+Supported platforms: `tiktok` | `instagram` | `facebook` | `youtube` | `x`
+
+```
+GET  /platform/get_oauth_account?agent_id={id}
+→ [ { platform, account_name, connected_at, ... } ]
+
+PUT  /platform/disconnect/{platform}
+Body: { "agent_id": "{id}" }
+→ 200
+```
+
+MCP tools: `gen_get_social_connect_url`, `gen_list_connected_socials`, `gen_disconnect_social`
+
 ## Publish to Social
 
-Currently TikTok. Instagram and YouTube coming.
+Supported platforms: **TikTok, Instagram, Facebook, YouTube, X**. Base URL: `https://python.gen.pro`
 
 ```
-POST /v1/user_jobs?agent_id={id}
+POST /schedule/with-post
 Content-Type: application/json
-Body:
-{
-  "user_job_type": "publish_content",
-  "data": "{\"platform\":\"tiktok\",\"media_url\":\"https://cdn.gen.pro/outputs/render_xyz.mp4\",\"description\":\"caption #hashtags\",\"schedule_type\":\"now\",\"media_type\":\"VIDEO\"}"
-}
-→ { user_job_id }
 ```
 
-Data fields (JSON-stringified in `data`):
+Fields:
 
 | Field | Type | Required | Notes |
 |-------|------|----------|-------|
-| `platform` | string | Yes | `tiktok` |
-| `media_url` | string | Yes | public URL to media (render URL works directly) |
-| `description` | string | Yes | caption, max ~2200 chars for TikTok; include hashtags inline |
-| `title` | string | No | post title |
-| `media_type` | string | No | `VIDEO` (default) \| `IMAGE` |
-| `schedule_type` | string | Yes | `now` \| `scheduled` |
-| `scheduled_time` | string | if `scheduled` | ISO 8601 UTC, must be future |
+| `agent_id` | string | Yes | agent posting the content |
+| `platform` | string[] | Yes | e.g. `["tiktok"]` or `["instagram"]` |
+| `schedule_type` | string | Yes | `now` \| `specific_time` |
+| `scheduled_time` | string | if `specific_time` | ISO 8601 UTC, must be future |
+| `description` | string | Yes for text/recommended | caption; include hashtags inline |
+| `media_url` | string | No | single video URL |
+| `media_urls` | string[] | No | image URLs (X: up to 4) |
+| `title` | string | No | post title (YouTube) |
+| `media_type` | string | No | `VIDEO` (default) \| `IMAGE` \| `TWEET_VIDEO` |
 | `thumbnail_url` | string | No | custom thumbnail URL |
-| `timezone_offset` | integer | No | minutes from UTC |
-
-### Upload media (upload_asset)
+| `timezone_offset` | number | No | UTC offset in hours (e.g. `-7` for PDT) |
+| `thread` | object[] | No | **X only** — `[{text, media_urls?}]` chain |
+| `reply_to_tweet_id` | string | No | **X only** — reply to this tweet id |
 
-Posts reference media by URL (public at post time). Upload a local file to get a hosted URL and save it to the agent's Assets.
+Post now (TikTok):
 
-```
-POST /v1/social/upload_asset   (multipart/form-data)
-Fields: file=<binary>, agent_id=<agent_id>
--> { "url": "https://cdn.gen.pro/...", "asset_id": "123", "content_resource_id": 123, "asset_registered": true }
+```json
+{
+  "agent_id": "<id>",
+  "platform": ["tiktok"],
+  "description": "caption #hashtags",
+  "media_url": "https://cdn.gen.pro/outputs/render_xyz.mp4",
+  "media_type": "VIDEO",
+  "schedule_type": "now"
+}
 ```
 
-Use the returned `url` as `media_url` or an entry in `media_urls`. Upload -> post:
+Schedule for later (Instagram):
 
-```bash
-URL=$(curl -s -X POST "https://api.gen.pro/v1/social/upload_asset" -H "X-API-Key: $KEY" \
-  -F "file=@image.jpg" -F "agent_id=$AGENT" | python3 -c "import sys,json;print(json.load(sys.stdin)['url'])")
-curl -X POST "https://api.gen.pro/v1/schedule/with-post" -H "X-API-Key: $KEY" -H "Content-Type: application/json" \
-  -d "{\"agent_id\":\"$AGENT\",\"platform\":[\"x\"],\"media_type\":\"IMAGE\",\"media_urls\":[\"$URL\"],\"schedule_type\":\"now\",\"description\":\"hi\"}"
+```json
+{
+  "agent_id": "<id>",
+  "platform": ["instagram"],
+  "description": "Morning routine 🌅 #wellness",
+  "media_url": "https://cdn.gen.pro/outputs/render_xyz.mp4",
+  "schedule_type": "specific_time",
+  "scheduled_time": "2026-08-01T09:00:00Z"
+}
 ```
 
-For up to 4 images on X, call upload_asset per file and pass all URLs in `media_urls`.
-
-### X (Twitter) posting
-
-X supports the richest post set via `POST /v1/schedule/with-post` (the `gen_schedule_post` tool). Beyond a single text/video tweet, X adds up to 4 images per tweet, threads, and replies.
+X thread:
 
-```
-POST /v1/schedule/with-post
-Content-Type: application/json
-Body:
+```json
 {
-  "agent_id": "<agent_id>",
+  "agent_id": "<id>",
   "platform": ["x"],
   "schedule_type": "now",
   "thread": [
@@ -1050,32 +1072,38 @@ Body:
 }
 ```
 
-X-specific fields on the schedule body:
+`thread` and `reply_to_tweet_id` are ignored for TikTok/Instagram/Facebook/YouTube. A thread that fails partway stops at the failed segment and is marked failed (not retried, to avoid duplicates).
 
-| Field | Type | Required | Notes |
-|-------|------|----------|-------|
-| `platform` | string[] | Yes | include `"x"` |
-| `description` | string | for text/reply | the tweet text |
-| `media_urls` | string[] | No | up to 4 image URLs for one tweet |
-| `media_type` | string | No | `TWEET_VIDEO` for video, `IMAGE` for images |
-| `thread` | object[] | No | **X only.** ordered list of `{text, media_urls?}` tweet segments; posted as a chain |
-| `reply_to_tweet_id` | string | No | **X only.** reply to an existing tweet id; combine with `thread` to reply-then-chain |
+### Upload media (upload_asset)
 
-`thread` and `reply_to_tweet_id` are ignored for TikTok/Instagram/Facebook/YouTube (single-media platforms). A thread that fails partway stops at the failed segment, records the tweets already posted, and is marked failed (not retried, to avoid duplicates).
+Upload a local file to get a hosted CDN URL for use in `media_url`/`media_urls`.
 
-### Poll the publish job
+```
+POST /v1/social/upload_asset   (multipart/form-data, api.gen.pro)
+Fields: file=<binary>, agent_id=<agent_id>
+→ { url, asset_id, content_resource_id, asset_registered }
+```
+
+For up to 4 images on X, call `upload_asset` per file and pass all URLs in `media_urls`.
+
+### Content calendar
 
 ```
-GET /v1/user_jobs/{id}?agent_id={id}
+GET    /schedule/posts?agent_id={id}[&start_date=2026-08-01&end_date=2026-08-31]
+GET    /schedule/post/{post_id}   → { id, status, platforms: [{platform, status, post_url, published_at}] }
+PATCH  /schedule/post/{post_id}   → update before it publishes (e.g. reschedule or edit caption)
+DELETE /schedule/post/{post_id}   → cancel a scheduled post
 ```
 
-Status flow: `pending → processing → completed | failed`. On `completed`, `result.post_id` is the platform's post ID.
+Status flow: `accepted → publishing → published | failed`. On `published`, `platforms[0].post_url` is the live link.
+
+MCP tools: `gen_schedule_post`, `gen_list_scheduled_posts`, `gen_get_post_status`, `gen_update_scheduled_post`, `gen_delete_scheduled_post`
 
 ### Prerequisites
 
-- Agent must have a connected TikTok account (connected via the GEN app)
-- Media must be publicly accessible at post time
-- For scheduled posts, `scheduled_time` must be in the future
+- Agent must have a connected account for the target platform (OAuth via `get_authorization_url`)
+- Media URL must be publicly accessible at post time
+- `scheduled_time` must be in the future for `specific_time` posts
 
 ---
 
@@ -1338,6 +1366,9 @@ delivery.
 - `POST /recurring-jobs/:id/pause` → `status = "paused"`, scheduler stops
   queueing new runs. History and configuration preserved.
 - `POST /recurring-jobs/:id/resume` → `status = "active"`.
+- `POST /recurring-jobs/:id/test-run` → fires one immediate execution now,
+  same path as a scheduled run. Returns `202 Accepted` with `run_id` to poll
+  via `GET /v1/agent/runs/:run_id`. Credits charged per normal execution.
 - `DELETE /recurring-jobs/:id` → `status = "deleted"`, returns 204. Use
   pause for reversible stops.
 
diff --git a/public/llms.txt b/public/llms.txt
index 030e323..73c8ad2 100644
--- a/public/llms.txt
+++ b/public/llms.txt
@@ -480,9 +480,9 @@ Only `ingredient` role columns can be created / updated / deleted.
 
 ## Step 5 — Export & Publish
 
-**What it is.** Composite all layers on the final_video cell into one MP4 (render), download it, optionally publish to TikTok (Instagram / YouTube coming).
+**What it is.** Composite all layers on the final_video cell into one MP4 (render), download it, and publish to social platforms: TikTok, Instagram, Facebook, YouTube, and X.
 
-**Mental model.** Render is an async generation like any other — trigger it on the final_video cell, poll until completed, then the `output_resources[0].url` is a public CDN URL. Publishing is a user_job that takes a `media_url` (the render URL works directly).
+**Mental model.** Render is an async generation — trigger it on the final_video cell, poll until completed, then the `output_resources[0].url` is a public CDN URL. Publishing takes a `media_url` (the render URL works directly). Connect social accounts first via the OAuth flow.
 
 **Render:**
 
@@ -492,25 +492,13 @@ curl -X POST "https://api.gen.pro/v1/vidsheet/$ENGINE_ID/cells/$FINAL_VIDEO_CELL
 # → { "generation_id": 56789, "status": "pending" }
 ```
 
-**Poll:**
+**Poll render:**
 
 ```bash
 curl "https://api.gen.pro/v1/generations/56789" \
   -H "X-API-Key: $GEN_API_KEY"
 # pending → processing → completed | failed
-# on completed:
-# {
-#   "id": 56789,
-#   "status": "completed",
-#   "output_resources": [
-#     {
-#       "id": 4821,
-#       "url": "https://cdn.gen.pro/outputs/render_xyz.mp4",
-#       "thumbnail_url": "https://cdn.gen.pro/thumbnails/render_xyz.jpg",
-#       "object_type": "video"
-#     }
-#   ]
-# }
+# on completed: output_resources[0].url is a public CDN URL
 ```
 
 **Download** (public CDN URL, no auth needed):
@@ -519,45 +507,70 @@ curl "https://api.gen.pro/v1/generations/56789" \
 curl -L -o final_video.mp4 "https://cdn.gen.pro/outputs/render_xyz.mp4"
 ```
 
-**Publish to TikTok now** (chain from the render URL):
+**Connect a social account** (OAuth URL — user opens in browser):
 
 ```bash
-curl -X POST "https://api.gen.pro/v1/user_jobs?agent_id=$AGENT_ID" \
+curl "https://python.gen.pro/platform/get_authorization_url?agent_id=$AGENT_ID&platform=tiktok" \
+  -H "X-API-Key: $GEN_API_KEY"
+# → { "url": "https://..." }   ← open this in a browser
+```
+
+**Publish now** (TikTok — chain from the render URL):
+
+```bash
+curl -X POST "https://python.gen.pro/schedule/with-post" \
   -H "X-API-Key: $GEN_API_KEY" \
   -H "Content-Type: application/json" \
   -d '{
-    "user_job_type": "publish_content",
-    "data": "{\"platform\":\"tiktok\",\"media_url\":\"https://cdn.gen.pro/outputs/render_xyz.mp4\",\"description\":\"The $2 taco at 2am #streetfood #tacotok\",\"schedule_type\":\"now\",\"media_type\":\"VIDEO\"}"
+    "agent_id": "'$AGENT_ID'",
+    "platform": ["tiktok"],
+    "media_url": "https://cdn.gen.pro/outputs/render_xyz.mp4",
+    "description": "The $2 taco at 2am #streetfood #tacotok",
+    "media_type": "VIDEO",
+    "schedule_type": "now"
   }'
-# → { "user_job_id": 138860 }
 ```
 
 **Schedule for later:**
 
 ```bash
-# same body but:
-# "schedule_type": "scheduled",
-# "scheduled_time": "2026-04-20T15:00:00Z"
+curl -X POST "https://python.gen.pro/schedule/with-post" \
+  -H "X-API-Key: $GEN_API_KEY" \
+  -H "Content-Type: application/json" \
+  -d '{
+    "agent_id": "'$AGENT_ID'",
+    "platform": ["instagram"],
+    "media_url": "https://cdn.gen.pro/outputs/render_xyz.mp4",
+    "description": "Morning routine 🌅 #wellness",
+    "schedule_type": "specific_time",
+    "scheduled_time": "2026-08-01T09:00:00Z"
+  }'
 ```
 
-**Poll publish job:**
+**Check publish status:**
 
 ```bash
-curl "https://api.gen.pro/v1/user_jobs/138860?agent_id=$AGENT_ID" \
+curl "https://python.gen.pro/schedule/post/POST_ID" \
   -H "X-API-Key: $GEN_API_KEY"
-# pending → processing → completed | failed
-# on completed: result.post_id is the TikTok post ID
+# status: accepted → publishing → published | failed
+# on published: platforms[0].post_url is the live link
 ```
 
-**Prerequisites for publish.** The agent must have a connected TikTok account. The media URL must be publicly accessible at post time.
+**Prerequisites for publish.** The agent must have a connected social account for the target platform (connected via OAuth). The media URL must be publicly accessible at post time.
 
 ### Step 5 endpoints
 
 ```
-POST   /v1/vidsheet/:id/cells/:cell_id/render?agent_id=   — kick off render
-GET    /v1/generations/:id                                          — poll render
-POST   /v1/user_jobs?agent_id= (publish_content)                    — publish now / schedule
-GET    /v1/user_jobs/:id?agent_id=                                  — poll publish job
+POST   /v1/vidsheet/:id/cells/:cell_id/render?agent_id=      — kick off render (api.gen.pro)
+GET    /v1/generations/:id                                   — poll render (api.gen.pro)
+GET    /platform/get_authorization_url                       — OAuth connect URL (python.gen.pro)
+GET    /platform/get_oauth_account                           — list connected socials (python.gen.pro)
+PUT    /platform/disconnect/:platform                        — disconnect social (python.gen.pro)
+POST   /schedule/with-post                                   — post or schedule (python.gen.pro)
+GET    /schedule/posts                                       — content calendar (python.gen.pro)
+GET    /schedule/post/:id                                    — poll post status (python.gen.pro)
+PATCH  /schedule/post/:id                                    — update scheduled post (python.gen.pro)
+DELETE /schedule/post/:id                                    — delete scheduled post (python.gen.pro)
 ```
 
 ---
diff --git a/src/content/docs/reference/cards/video-from-image.mdx b/src/content/docs/reference/cards/video-from-image.mdx
index b16e404..29242d6 100644
--- a/src/content/docs/reference/cards/video-from-image.mdx
+++ b/src/content/docs/reference/cards/video-from-image.mdx
@@ -44,6 +44,7 @@ POST /v1/vidsheet/{id}/cells/{cell_id}/generate?agent_id={agent_id}
 | Seedance Lite | `video_from_image` | `"seedance_lite"` |
 | Seedance Pro | `video_from_image` | `"seedance_pro"` |
 | Seedance Pro 1.5 | `video_from_image` | `"seedance_pro_1_5"` |
+| Seedance 2.0 | `video_from_image` | `"seedance-2.0"` |
 | Grok | `video_from_image` | `"grok"` |
 
 ---
diff --git a/src/content/docs/reference/cards/video-from-text.mdx b/src/content/docs/reference/cards/video-from-text.mdx
index d65bd83..242af38 100644
--- a/src/content/docs/reference/cards/video-from-text.mdx
+++ b/src/content/docs/reference/cards/video-from-text.mdx
@@ -44,6 +44,7 @@ POST /v1/vidsheet/{id}/cells/{cell_id}/generate?agent_id={agent_id}
 | Kling 1.6 | `video_from_text` | `"kling_1_6"` |
 | Seedance Pro | `video_from_text` | `"seedance_pro"` |
 | Seedance Pro 1.5 | `video_from_text` | `"seedance_pro_1_5"` |
+| Seedance 2.0 | `video_from_text` | `"seedance-2.0"` |
 | Grok | `video_from_text` | `"grok"` |
 
 ---
@@ -55,7 +56,7 @@ POST /v1/vidsheet/{id}/cells/{cell_id}/generate?agent_id={agent_id}
 | `generation_type` | string | Yes | `"video_from_text"`. |
 | `data.prompt` | string | Yes | Text description of the video to generate. Supports `{{variable_name}}` syntax. |
 | `data.negative_prompt` | string | No | What to avoid in the generated video. |
-| `data.model` | string | No | Model variant. See [Models](#models) table. |
+| `data.model` | string | No | Model variant — one of `"veo_3"`, `"veo_3_fast"`, `"veo_3_1"`, `"veo_3_1_fast"`, `"sora_2"`, `"kling_1_6"`, `"seedance_pro"`, `"seedance_pro_1_5"`, `"seedance-2.0"`, `"grok"`. See [Models](#models) table. |
 | `data.aspect_ratio` | string | No | `"1:1"`, `"9:16"`, or `"16:9"`. Grok also supports `"4:3"`, `"3:4"`, `"3:2"`, and `"2:3"`. |
 | `data.resolution` | string | No | Output resolution. Use `"1080p"` or `"720p"` when supported by the selected model. If omitted, the API uses the cell default or the closest supported model default. |
 | `data.duration` | integer | No | Video duration in seconds. |
diff --git a/src/content/docs/reference/publishing.mdx b/src/content/docs/reference/publishing.mdx
index 9245093..3b8c132 100644
--- a/src/content/docs/reference/publishing.mdx
+++ b/src/content/docs/reference/publishing.mdx
@@ -1,228 +1,138 @@
 ---
 title: Publishing
-description: Post and schedule content to social media platforms.
+description: Connect social accounts and post or schedule content to TikTok, Instagram, Facebook, YouTube, and X.
 ---
 
 import { Aside } from '@astrojs/starlight/components';
 
-Publishing lets you post content to social media platforms immediately or schedule it for later. Currently supports TikTok, with more platforms coming soon.
+Publishing lets you connect social accounts and post content immediately or schedule it for later. Supported platforms: **TikTok, Instagram, Facebook, YouTube, and X**.
 
 <Aside type="note">
-  Publishing requires authentication via the `X-API-Key` header. The agent must have a connected TikTok social account (via AURA) before publishing.
+  Publishing requires authentication via the `X-API-Key` header. Connect a social account first via the OAuth flow before posting.
 </Aside>
 
-## Publish content
+## Connect a social account
 
-Posts or schedules content to a social media platform. Uses the standard user job creation endpoint.
+### Get the OAuth URL
+
+Returns a single-use OAuth URL the user opens in a browser to authorize GEN to post on their behalf. Never ask for credentials directly — always use this flow.
 
 ```
-POST /v1/user_jobs?agent_id={agent_id}
+GET /platform/get_authorization_url
 ```
 
-### Parameters
+Base URL: `https://python.gen.pro`
+
+#### Parameters
 
 | Parameter | Type | Required | Description |
 |-----------|------|----------|-------------|
-| `agent_id` | string | Yes | The agent ID (query parameter). |
-| `user_job_type` | string | Yes | Must be `publish_content`. |
-| `data` | string (JSON) | Yes | JSON-stringified object containing the post configuration (see Data fields below). |
-
-### Data fields
-
-The `data` value is a JSON string containing:
-
-| Field | Type | Required | Description |
-|-------|------|----------|-------------|
-| `platform` | string | Yes | Target platform. Currently `tiktok`. |
-| `media_url` | string | Yes | Public URL to the media file (S3 presigned URL, CDN link, etc). Must be accessible at post time. |
-| `description` | string | Yes | Post caption/description. Max ~2200 characters for TikTok. Include hashtags inline. |
-| `title` | string | No | Post title (used on some platforms). |
-| `media_type` | string | No | `VIDEO` (default) or `IMAGE`. |
-| `schedule_type` | string | Yes | `now` for immediate posting. `scheduled` for future posting. |
-| `scheduled_time` | string | Conditional | ISO 8601 UTC datetime (e.g. `2026-03-16T15:00:00Z`). Required when `schedule_type` is `scheduled`. Must be in the future. |
-| `thumbnail_url` | string | No | Custom thumbnail URL. Auto-generated from video if not provided. |
-| `timezone_offset` | integer | No | Timezone offset in minutes from UTC. Default `0`. |
+| `agent_id` | string | Yes | The agent to attach the account to. |
+| `platform` | string | Yes | `tiktok` \| `instagram` \| `facebook` \| `youtube` \| `x` |
 
-### Schedule types
-
-| Value | Behavior |
-|-------|----------|
-| `now` | Posts immediately via the connected social account. |
-| `scheduled` | Queues for posting at `scheduled_time`. |
-
-### Response
-
-```json
-{
-  "user_job_id": 138860
-}
-```
-
-### Status flow
-
-Poll status via `GET /v1/user_jobs/{id}?agent_id={agent_id}`.
-
-| Status | Description |
-|--------|-------------|
-| `pending` | Job created, queued for processing. |
-| `processing` | Post is being submitted to the platform. |
-| `completed` | Post published successfully. The `result` field contains the `post_id` from the platform. |
-| `failed` | Publishing failed. Check `failed_reason`. |
-
-### Example: Post now
+#### Example
 
 ```bash
-curl -X POST "https://api.gen.pro/v1/user_jobs?agent_id=your-agent-id" \
-  -H "X-API-Key: your-api-key" \
-  -H "Content-Type: application/json" \
-  -d '{
-    "user_job_type": "publish_content",
-    "data": "{\"platform\": \"tiktok\", \"media_url\": \"https://cdn.example.com/video.mp4\", \"description\": \"Factory price reveal #sourcing #china\", \"schedule_type\": \"now\", \"media_type\": \"VIDEO\"}"
-  }'
+curl "https://python.gen.pro/platform/get_authorization_url?agent_id=your-agent-id&platform=tiktok" \
+  -H "X-API-Key: your-api-key"
 ```
 
-### Example: Schedule for later
-
-```bash
-curl -X POST "https://api.gen.pro/v1/user_jobs?agent_id=your-agent-id" \
-  -H "X-API-Key: your-api-key" \
-  -H "Content-Type: application/json" \
-  -d '{
-    "user_job_type": "publish_content",
-    "data": "{\"platform\": \"tiktok\", \"media_url\": \"https://cdn.example.com/video.mp4\", \"description\": \"Your $12 beauty sponge costs 10 cents #beauty #factory\", \"title\": \"Beauty sponge price reveal\", \"schedule_type\": \"scheduled\", \"scheduled_time\": \"2026-03-16T15:00:00Z\", \"media_type\": \"VIDEO\"}"
-  }'
-```
-
-### Prerequisites
-
-- Agent must have a connected TikTok social account (connected via AURA)
-- Video/image must be publicly accessible via URL at post time
-- For scheduled posts, `scheduled_time` must be in the future
-
-## Upload media
-
-Posts reference media by **URL** — the media must be publicly fetchable at post time. Upload a local file with `upload_asset` to get a durable, hosted URL you can pass to `media_url` / `media_urls`. The uploaded file is also saved to the agent's Assets, so you can reuse it later.
+### List connected accounts
 
 ```
-POST /v1/social/upload_asset
+GET /platform/get_oauth_account?agent_id={agent_id}
 ```
 
-Send the file as `multipart/form-data`.
-
-### Parameters
-
-| Parameter | Type | Required | Description |
-|-----------|------|----------|-------------|
-| `file` | file | Yes | The video or image file (multipart). |
-| `agent_id` | string | Yes | The agent whose Assets the file is saved to. |
-
-### Response
+Base URL: `https://python.gen.pro`
 
-```json
-{
-  "url": "https://cdn.gen.pro/....jpg",
-  "cdn_url": "https://cdn.gen.pro/....jpg",
-  "asset_id": "12345",
-  "content_resource_id": 12345,
-  "asset_registered": true
-}
+```bash
+curl "https://python.gen.pro/platform/get_oauth_account?agent_id=your-agent-id" \
+  -H "X-API-Key: your-api-key"
 ```
 
-Use the returned `url` as `media_url` (single) or inside `media_urls` when scheduling a post.
-
-### Example
+### Disconnect an account
 
-```bash
-curl -X POST "https://api.gen.pro/v1/social/upload_asset" \
-  -H "X-API-Key: your-api-key" \
-  -F "file=@/path/to/image.jpg" \
-  -F "agent_id=your-agent-id"
+```
+PUT /platform/disconnect/:platform
 ```
 
-### Upload → post recipe
+Base URL: `https://python.gen.pro`
 
 ```bash
-# 1. Upload the file, capture the hosted URL
-URL=$(curl -s -X POST "https://api.gen.pro/v1/social/upload_asset" \
-  -H "X-API-Key: your-api-key" \
-  -F "file=@/path/to/image.jpg" \
-  -F "agent_id=your-agent-id" | python3 -c "import sys,json;print(json.load(sys.stdin)['url'])")
-
-# 2. Post it (X example — a tweet with one image)
-curl -X POST "https://api.gen.pro/v1/schedule/with-post" \
+curl -X PUT "https://python.gen.pro/platform/disconnect/tiktok" \
   -H "X-API-Key: your-api-key" \
   -H "Content-Type: application/json" \
-  -d "{\"agent_id\":\"your-agent-id\",\"platform\":[\"x\"],\"description\":\"from the API\",\"media_type\":\"IMAGE\",\"media_urls\":[\"$URL\"],\"schedule_type\":\"now\"}"
+  -d '{"agent_id": "your-agent-id"}'
 ```
 
-For up to 4 images on X, call `upload_asset` per file and pass all the URLs in `media_urls`.
-
-## Posting to X (Twitter)
+---
 
-X supports the richest set of post types: a plain text tweet, a tweet with up to **4 images**, a tweet with a video, a **thread** (an ordered chain of tweets), and a **reply** to an existing tweet. X posts are created through the scheduling API.
+## Schedule a post
 
 ```
-POST /v1/schedule/with-post
+POST /schedule/with-post
 ```
 
+Base URL: `https://python.gen.pro`
+
 ### Fields
 
 | Field | Type | Required | Description |
 |-------|------|----------|-------------|
-| `agent_id` | string | Yes | The agent whose connected X account posts. |
-| `platform` | string[] | Yes | Include `"x"`. |
-| `schedule_type` | string | Yes | `now`, `specific_time`, or `next_available`. |
-| `scheduled_time` | string | Conditional | Required when `schedule_type` is `specific_time`. |
-| `description` | string | Conditional | The tweet text. Required for a text-only tweet or reply. |
-| `media_type` | string | No | `TWEET_VIDEO` for video, `IMAGE` for images. |
-| `media_url` | string | No | A single video URL. |
-| `media_urls` | string[] | No | Up to **4 image URLs** for a single tweet. |
-| `thread` | object[] | No | **X only.** An ordered list of tweet segments to post as a chain. Each segment is `{ "text": "...", "media_urls": ["..."] }` (per-segment images optional, up to 4). |
-| `reply_to_tweet_id` | string | No | **X only.** Post as a reply to this tweet id. Combine with `thread` to reply and then continue a chain. |
+| `agent_id` | string | Yes | The agent posting. |
+| `platform` | string[] | Yes | Array with the target platform: `["tiktok"]`, `["instagram"]`, etc. |
+| `schedule_type` | string | Yes | `now` (immediate) or `specific_time` (requires `scheduled_time`). |
+| `description` | string | Conditional | Post caption. Required for text-only posts (X) or recommended for all. |
+| `media_url` | string | No | Single video URL. |
+| `media_urls` | string[] | No | Image URLs (X supports up to 4). |
+| `title` | string | No | Post title (YouTube). |
+| `media_type` | string | No | `VIDEO` (default) \| `IMAGE` \| `TWEET_VIDEO` |
+| `scheduled_time` | string | Conditional | ISO 8601 datetime — required when `schedule_type` is `specific_time`. |
+| `thumbnail_url` | string | No | Custom thumbnail URL. |
+| `timezone_offset` | number | No | Local UTC offset in hours (e.g. `-7` for PDT). Omit to use the agent's configured timezone. |
+| `thread` | object[] | No | **X only.** Ordered chain of tweets: `[{"text":"..."}, {"text":"...", "media_urls":["..."]}]` |
+| `reply_to_tweet_id` | string | No | **X only.** Post as a reply to this tweet id. |
 
 <Aside type="note">
-  `thread` and `reply_to_tweet_id` apply to **X only**. They are ignored for TikTok, Instagram, Facebook, and YouTube, which post a single piece of media.
+  `thread` and `reply_to_tweet_id` apply to **X only**.
 </Aside>
 
-### Example: Text-only tweet
+### Example: Post now (TikTok)
 
 ```bash
-curl -X POST "https://api.gen.pro/v1/schedule/with-post" \
+curl -X POST "https://python.gen.pro/schedule/with-post" \
   -H "X-API-Key: your-api-key" \
   -H "Content-Type: application/json" \
   -d '{
     "agent_id": "your-agent-id",
-    "platform": ["x"],
-    "description": "gm from the API 👋",
+    "platform": ["tiktok"],
+    "description": "Factory price reveal #sourcing #china",
+    "media_url": "https://cdn.example.com/video.mp4",
+    "media_type": "VIDEO",
     "schedule_type": "now"
   }'
 ```
 
-### Example: Tweet with up to 4 images
+### Example: Scheduled post (Instagram)
 
 ```bash
-curl -X POST "https://api.gen.pro/v1/schedule/with-post" \
+curl -X POST "https://python.gen.pro/schedule/with-post" \
   -H "X-API-Key: your-api-key" \
   -H "Content-Type: application/json" \
   -d '{
     "agent_id": "your-agent-id",
-    "platform": ["x"],
-    "description": "four pics",
-    "media_type": "IMAGE",
-    "media_urls": [
-      "https://cdn.example.com/1.jpg",
-      "https://cdn.example.com/2.jpg",
-      "https://cdn.example.com/3.jpg",
-      "https://cdn.example.com/4.jpg"
-    ],
-    "schedule_type": "now"
+    "platform": ["instagram"],
+    "description": "Morning routine 🌅 #wellness",
+    "media_url": "https://cdn.example.com/video.mp4",
+    "schedule_type": "specific_time",
+    "scheduled_time": "2026-08-01T09:00:00Z"
   }'
 ```
 
-### Example: Thread
+### Example: X thread
 
 ```bash
-curl -X POST "https://api.gen.pro/v1/schedule/with-post" \
+curl -X POST "https://python.gen.pro/schedule/with-post" \
   -H "X-API-Key: your-api-key" \
   -H "Content-Type: application/json" \
   -d '{
@@ -230,35 +140,98 @@ curl -X POST "https://api.gen.pro/v1/schedule/with-post" \
     "platform": ["x"],
     "schedule_type": "now",
     "thread": [
-      { "text": "1/ Here is a thread about our launch." },
-      { "text": "2/ With a screenshot", "media_urls": ["https://cdn.example.com/shot.jpg"] },
-      { "text": "3/ Wrapping up. Thanks for reading!" }
+      { "text": "1/ Our launch thread:" },
+      { "text": "2/ Here is what we built", "media_urls": ["https://cdn.example.com/shot.jpg"] },
+      { "text": "3/ Thanks for reading!" }
     ]
   }'
 ```
 
-### Example: Reply to a tweet
+---
+
+## Content calendar
+
+### List scheduled posts
+
+```
+GET /schedule/posts?agent_id={agent_id}
+```
+
+Optional: `start_date` and `end_date` (ISO 8601) to filter the calendar window.
+
+```bash
+curl "https://python.gen.pro/schedule/posts?agent_id=your-agent-id&start_date=2026-08-01&end_date=2026-08-31" \
+  -H "X-API-Key: your-api-key"
+```
+
+### Get post status
+
+Check whether a post actually went live (not just queued). Returns platform-confirmed status.
+
+```
+GET /schedule/post/:post_id
+```
+
+```bash
+curl "https://python.gen.pro/schedule/post/post-id-here" \
+  -H "X-API-Key: your-api-key"
+```
+
+#### Response
+
+```json
+{
+  "id": "post-id",
+  "status": "published",
+  "platforms": [
+    {
+      "platform": "tiktok",
+      "status": "published",
+      "platform_post_id": "7123456789",
+      "post_url": "https://tiktok.com/@user/video/7123456789",
+      "published_at": "2026-08-01T09:01:12Z",
+      "error": null
+    }
+  ]
+}
+```
+
+Overall status flow: `accepted` → `publishing` → `published` | `failed`.
+
+### Update a scheduled post
+
+Edit a post before it goes out. Only works on posts that have not yet been published.
+
+```
+PATCH /schedule/post/:post_id
+```
 
 ```bash
-curl -X POST "https://api.gen.pro/v1/schedule/with-post" \
+curl -X PATCH "https://python.gen.pro/schedule/post/post-id-here" \
   -H "X-API-Key: your-api-key" \
   -H "Content-Type: application/json" \
-  -d '{
-    "agent_id": "your-agent-id",
-    "platform": ["x"],
-    "description": "Great point — here is our take.",
-    "reply_to_tweet_id": "1780000000000000000",
-    "schedule_type": "now"
-  }'
+  -d '{"scheduled_time": "2026-08-02T10:00:00Z"}'
+```
+
+### Delete a scheduled post
+
+```
+DELETE /schedule/post/:post_id
+```
+
+```bash
+curl -X DELETE "https://python.gen.pro/schedule/post/post-id-here" \
+  -H "X-API-Key: your-api-key"
 ```
 
-If a thread fails partway through, posting stops at the failed segment; the tweets already posted are recorded and the post is marked failed (it is not retried, to avoid duplicate tweets).
+---
 
-### Errors
+## Errors
 
 | Status | Error code | Description |
 |--------|------------|-------------|
 | `401` | `unauthorized` | Missing or invalid API key. |
 | `403` | `permission_denied` | You do not have access to this agent. |
 | `422` | `validation_error` | Invalid or missing required fields. |
-| `422` | `no_active_credit_purchase` | The agent's workspace has no active credit purchase. |
+| `422` | `insufficient_credits` | The agent's workspace has no usable credits. |
+| `404` | `not_found` | The post or agent was not found. |
diff --git a/src/content/docs/reference/recurring-jobs.mdx b/src/content/docs/reference/recurring-jobs.mdx
index dad4bde..56b7bc6 100644
--- a/src/content/docs/reference/recurring-jobs.mdx
+++ b/src/content/docs/reference/recurring-jobs.mdx
@@ -181,6 +181,23 @@ Returns `204 No Content` on success.
 }
 ```
 
+## Run a job immediately
+
+Trigger one execution right now without waiting for the scheduled time. Fires the same execution path as a scheduled run — credits are charged and the run is recorded. Useful for testing a new job before its first scheduled firing.
+
+```
+POST /v1/agents/:agent_id/recurring-jobs/:job_id/test-run
+```
+
+Base URL: `https://agent.gen.pro`
+
+```bash
+curl -X POST "https://agent.gen.pro/v1/agents/your-agent-id/recurring-jobs/your-job-id/test-run" \
+  -H "X-API-Key: your-api-key"
+```
+
+Returns `202 Accepted` with a `run_id` you can poll via `GET /v1/agent/runs/:run_id`.
+
 ## Errors
 
 | Status | Code | Description |

From 86544e15c7293d0312c34984398c4622c00717eb Mon Sep 17 00:00:00 2001
From: mavneox <mav@onlywithneo.com>
Date: Wed, 29 Jul 2026 09:17:56 +0000
Subject: [PATCH 5/7] docs: add MCP tool name annotations to llms-full.txt
 endpoint sections

10 targeted additions linking MCP tool names to existing endpoint blocks
that lacked them: avatars, song mixes, conversations, content ideas,
research, content monitoring, generations, direct uploads, watchlists,
and recurring jobs.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
---
 public/llms-full.txt | 45 ++++++++++++++++++++++++++++++++++++++++----
 1 file changed, 41 insertions(+), 4 deletions(-)

diff --git a/public/llms-full.txt b/public/llms-full.txt
index d42e497..c9581b8 100644
--- a/public/llms-full.txt
+++ b/public/llms-full.txt
@@ -79,6 +79,15 @@ id may be underscore-joined for multi-delete (e.g. "7_8_9").
 → 200 empty
 ```
 
+MCP shortcuts for avatar management:
+- `gen_list_agent_avatars(agent_id)` — list avatars for an agent
+- `gen_create_agent_avatar(agent_id, degod_avatar_id)` — attach a DeGod collection avatar
+- `gen_delete_agent_avatar(agent_id, avatar_id)` — delete one or more avatars
+- `gen_add_agent_look(agent_id, signed_id)` — attach an uploaded image as a non-primary look (after `gen_create_direct_upload`)
+- `gen_set_agent_primary_photo(agent_id, signed_id)` — upload and set as primary avatar (replaces existing primary)
+- `gen_add_agent_account(agent_id, url, platform?)` — add one of the agent's OWN social accounts (what it posts from)
+- `gen_add_agent_inspiration(agent_id, url)` — add one inspiration source (creator the agent draws style from)
+
 ## Agent Core (preferred single-call setup)
 
 **Use this for all agent setup reads and writes.** Flat endpoint. Every field name mirrors the GEN Setup canvas.
@@ -362,6 +371,8 @@ Live progress:
 
 Backend/API-first contract for combining multiple songs into one long DJ-style audio output.
 
+MCP tool: `gen_create_song_mix(agent_id, tracks?, prompt?, transition_style?, output_format?, auto_order?, desired_duration_seconds?, generate_missing_count?)` — combine existing songs into a DJ-style mix. Returns `status=pending_engine` until the renderer worker is installed, or `pending_song_generation` when source songs must be generated first.
+
 ```
 POST /v1/audio/mixes
 Body: {
@@ -416,6 +427,8 @@ PATCH  /v1/agent/conversations/{id}                    — Body: { title?, pinne
 DELETE /v1/agent/conversations/{id}                    — deletes conversation, runs, messages
 ```
 
+MCP tools: `gen_list_conversations(agent_id)`, `gen_get_conversation(agent_id, conversation_id)`, `gen_get_session_preferences(agent_id, conversation_id)`, `gen_update_session_preferences(agent_id, conversation_id, preferences)`.
+
 ## Content Ideas
 
 Ideas produced by runs are first-class records.
@@ -451,6 +464,8 @@ Edit/rejection statuses: `change_idea`, `change_video`, `rejected`.
 
 Video type IDs: `0=Any, 1=Talking Avatar, 2=Green Screen, 3=Montage, 4=Text-Driven, 5=POV Object, 6=Narrated/VO, 8=Split Screen, 9=Skit`.
 
+MCP tools: `gen_list_content_ideas(agent_id, status?)`, `gen_update_idea_status(agent_id, idea_id, status)`, `gen_expand_idea(agent_id, idea_id)`, `gen_refine_content_ideas(agent_id, conversation_id, feedback)`, `gen_decide_agent_run(agent_id, run_id, approved)`, `gen_set_content_preference(agent_id, preference)`.
+
 ## Research (standalone)
 
 ```
@@ -463,6 +478,8 @@ Poll `GET /v1/agent/runs/{run_id}` until `status` is `completed`. Completed mess
 
 Depths: `quick`, `default`, `deep`.
 
+MCP tool: `gen_run_research(agent_id, topic, depth?)` — returns a `run_id` to poll with `gen_get_run_status`.
+
 ## Research topics (daily background research)
 
 Configured via Agent Core (`research_topics` field). The platform researches each topic daily and feeds findings into the Ideas Engine automatically. No poll needed — just set topics and consume ideas downstream.
@@ -512,6 +529,8 @@ No raw data dump endpoint — query through the agent:
 
 Status flow: `pending → processing → completed` (one-time) or stays `processing` (monitoring). Poll `GET /v1/user_jobs/{id}?agent_id={id}`.
 
+MCP tools: `gen_create_monitoring_job(agent_id, platform, search_type, value, days?, country?, max_results?, monitoring?, comment_monitoring?)`, `gen_update_monitoring_job(agent_id, job_id, ...)` — only pending/processing jobs can be updated.
+
 ---
 
 # Phase 3 — Convert
@@ -630,6 +649,13 @@ POST /v1/generations/{id}/stop       — credits refunded; status → "stopped"
 POST /v1/generations/{id}/continue   — credits re-charged; status → "pending"
 ```
 
+MCP tools:
+- `gen_generate_content(agent_id, sheet_id, cell_id, generation_type, data)` — THE WORKHORSE. Trigger cell generation. Returns `generation_id` to poll with `gen_get_generation`.
+- `gen_generate_layer(agent_id, sheet_id, cell_id, layer_id)` — trigger generation for a specific layer (uses the layer's already-configured generation type and data).
+- `gen_get_generation(generation_id)` — poll until `status == "completed"`, then read `output_resources`.
+- `gen_stop_generation(generation_id)` — stop and refund credits.
+- `gen_continue_generation(generation_id)` — resume a stopped generation (credits re-charged).
+
 Status flow: `pending → processing → completed | failed | stopped`.
 
 ## Creation Card Types (exact `generation_type` + `data`)
@@ -789,6 +815,14 @@ Workflow:
 Compute checksum: openssl md5 -binary file | base64
 ```
 
+MCP tools for asset management:
+- `gen_create_direct_upload(agent_id, filename, byte_size, checksum, content_type)` — returns `signed_id` + upload URL
+- `gen_complete_asset_upload(agent_id, signed_id)` — attach an already-uploaded blob to the agent's durable asset library; returns `content_resource` id and CDN URL
+- `gen_create_content_resource(agent_id, signed_id)` — create a content resource from a signed_id (alias for the POST endpoint above)
+- `gen_list_content_resources(agent_id, type?)` — list content resources with optional type filter
+- `gen_get_content_resource(agent_id, id)` — get a single resource with its public URL and generator info
+- `gen_delete_content_resource(agent_id, id)` — permanently delete a content resource
+
 ## Proof of Genesis / Backup to Blockchain (`https://agent-core.gen.pro/v1`)
 
 Proof of Genesis backs up assets from the GEN assets page or vidsheet workflows to Walrus and records proof metadata. It supports manual add/remove and automatic backup of selected asset events.
@@ -1228,6 +1262,7 @@ DELETE /v1/agents/{agent_id}/watchlists/{watchlist_id}/sources?platform=&target_
   restoring it if it had been soft-deleted.
 - **`PATCH /watchlists/:id` with `{"intent_active": false}`** is the pause
   primitive; `true` resumes. Use `DELETE` only to permanently retire.
+  MCP: `gen_pause_watchlist(agent_id, watchlist_id)` / `gen_resume_watchlist(agent_id, watchlist_id)`.
 - **`DELETE`** is a soft delete: rows are marked inactive and `deleted_at`
   is set; nothing is returned by subsequent `GET`/`list`.
 
@@ -1364,13 +1399,15 @@ delivery.
 ### Pause / resume / delete
 
 - `POST /recurring-jobs/:id/pause` → `status = "paused"`, scheduler stops
-  queueing new runs. History and configuration preserved.
-- `POST /recurring-jobs/:id/resume` → `status = "active"`.
+  queueing new runs. History and configuration preserved. MCP: `gen_pause_recurring_job(agent_id, job_id)`
+- `POST /recurring-jobs/:id/resume` → `status = "active"`. MCP: `gen_resume_recurring_job(agent_id, job_id)`
 - `POST /recurring-jobs/:id/test-run` → fires one immediate execution now,
   same path as a scheduled run. Returns `202 Accepted` with `run_id` to poll
-  via `GET /v1/agent/runs/:run_id`. Credits charged per normal execution.
+  via `GET /v1/agent/runs/:run_id`. Credits charged per normal execution. MCP: `gen_run_recurring_job_now(agent_id, job_id)`
 - `DELETE /recurring-jobs/:id` → `status = "deleted"`, returns 204. Use
-  pause for reversible stops.
+  pause for reversible stops. MCP: `gen_delete_recurring_job(agent_id, job_id)`
+- Get a single job: `gen_get_recurring_job(agent_id, job_id)` — returns 404 if deleted.
+- Idempotent default setup: `gen_ensure_default_recurring_job(agent_id)` — creates the default daily content-ideas job if none exists.
 
 ### Credit semantics
 

From a3308006eaee7949750b09018e4ffdbc5b0dc16a Mon Sep 17 00:00:00 2001
From: mavneox <mav@onlywithneo.com>
Date: Thu, 30 Jul 2026 04:16:05 -0500
Subject: [PATCH 6/7] docs: add seedance-2.0 and new kling models to video
 generation model lists

Update video_from_text and video_from_image model enumerations across
llms.txt, llms-full.txt, and generation-types.mdx to include the full
current model set: seedance-2.0, seedance-1.5-pro, seedance-1.0-pro,
seedance-1.0-lite, kling_2_1, kling_2_6 (previously missing from
video_from_text list).

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
---
 public/llms-full.txt                            | 6 +++---
 public/llms.txt                                 | 4 ++--
 src/content/docs/reference/generation-types.mdx | 4 ++--
 3 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/public/llms-full.txt b/public/llms-full.txt
index c9581b8..e726b7d 100644
--- a/public/llms-full.txt
+++ b/public/llms-full.txt
@@ -683,9 +683,9 @@ Output: image content resource
 
 ```
 generation_type: "video_from_text"
-Models: veo_3, veo_3_fast, veo_3_1, veo_3_1_fast, sora_2, kling_1_6, seedance_pro, seedance_pro_1_5, grok
+Models: veo_3, veo_3_fast, veo_3_1, veo_3_1_fast, sora_2, kling_1_6, kling_2_1, kling_2_6, seedance-2.0, seedance-1.5-pro, seedance-1.0-pro, seedance-1.0-lite, seedance_pro, seedance_pro_1_5, grok
 Aspect ratios: "1:1", "9:16", "16:9"; grok also supports "4:3", "3:4", "3:2", "2:3"
-Durations: model-dependent
+Durations: model-dependent (seedance-2.0 supports 4-15s)
 data: { prompt, negative_prompt?, model, aspect_ratio, resolution?, duration, native_audio?, variables }
 Resolution values: `1080p` or `720p` when supported by the selected model. Native audio is optional and defaults to false when omitted.
 Output: video content resource
@@ -695,7 +695,7 @@ Output: video content resource
 
 ```
 generation_type: "video_from_image"
-Models: kling_2_1, kling_2_6, veo_3, veo_3_fast, veo_3_1, veo_3_1_fast, sora_2, seedance_lite, seedance_pro, seedance_pro_1_5, grok
+Models: kling_2_1, kling_2_6, veo_3, veo_3_fast, veo_3_1, veo_3_1_fast, sora_2, seedance-2.0, seedance_lite, seedance_pro, seedance_pro_1_5, grok
 data: { prompt, negative_prompt?, model, image_resource_id, image_tail_resource_id?, aspect_ratio, resolution?, duration, native_audio? }
 Resolution values: `1080p` or `720p` when supported by the selected model. Native audio is optional and defaults to false when omitted.
 Output: video content resource
diff --git a/public/llms.txt b/public/llms.txt
index 73c8ad2..828facf 100644
--- a/public/llms.txt
+++ b/public/llms.txt
@@ -612,14 +612,14 @@ Models: `gemini_image`, `gemini_pro_image`, `midjourney`, `grok`. Aspect ratios:
 { "generation_type": "video_from_text",
   "data": { "prompt": "...", "model": "veo_3", "aspect_ratio": "9:16", "resolution": "1080p", "duration": 10, "native_audio": false, "negative_prompt": "no text overlays" } }
 ```
-Models: `veo_3`, `veo_3_fast`, `veo_3_1`, `veo_3_1_fast`, `sora_2`, `kling_1_6`, `seedance_pro`, `seedance_pro_1_5`, `grok`. Aspect ratios: `1:1`, `9:16`, `16:9`; Grok also supports `4:3`, `3:4`, `3:2`, `2:3`. Duration and resolution are model-dependent; use `resolution: "1080p"` or `"720p"` where supported. `native_audio` is optional and defaults to `false`.
+Models: `veo_3`, `veo_3_fast`, `veo_3_1`, `veo_3_1_fast`, `sora_2`, `kling_1_6`, `kling_2_1`, `kling_2_6`, `seedance-2.0`, `seedance-1.5-pro`, `seedance-1.0-pro`, `seedance-1.0-lite`, `seedance_pro`, `seedance_pro_1_5`, `grok`. Aspect ratios: `1:1`, `9:16`, `16:9`; Grok also supports `4:3`, `3:4`, `3:2`, `2:3`. Duration and resolution are model-dependent (seedance-2.0 supports 4-15s); use `resolution: "1080p"` or `"720p"` where supported. `native_audio` is optional and defaults to `false`.
 
 ### video_from_image
 ```json
 { "generation_type": "video_from_image",
   "data": { "image_resource_id": 4821, "image_tail_resource_id": 4822, "prompt": "...", "model": "kling_2_6", "aspect_ratio": "9:16", "resolution": "1080p", "duration": 5, "native_audio": false } }
 ```
-Models: `kling_2_1`, `kling_2_6`, `veo_3`, `veo_3_1`, `sora_2`, `seedance_lite`, `seedance_pro`, `seedance_pro_1_5`, `grok`. `image_tail_resource_id` optional. `resolution` and `native_audio` are optional and model-dependent.
+Models: `kling_2_1`, `kling_2_6`, `veo_3`, `veo_3_1`, `sora_2`, `seedance-2.0`, `seedance_lite`, `seedance_pro`, `seedance_pro_1_5`, `grok`. `image_tail_resource_id` optional. `resolution` and `native_audio` are optional and model-dependent.
 
 ### video_from_ingredients
 ```json
diff --git a/src/content/docs/reference/generation-types.mdx b/src/content/docs/reference/generation-types.mdx
index 43ba64d..78b9dc7 100644
--- a/src/content/docs/reference/generation-types.mdx
+++ b/src/content/docs/reference/generation-types.mdx
@@ -17,8 +17,8 @@ Every cell or layer generation declares a `generation_type` and a `data` object.
 |---|---|---|
 | `text` | text | gemini_2_0_flash, gemini_2_5_pro, gpt_4o, gpt_4o_mini, o3_mini, o4_mini, claude_sonnet_4 |
 | `image_from_text` | image | gemini_image, gemini_pro_image, midjourney, grok |
-| `video_from_text` | video | veo_3, veo_3_fast, veo_3_1, veo_3_1_fast, sora_2, kling_1_6, seedance_pro, seedance_pro_1_5, grok |
-| `video_from_image` | video | kling_2_1, kling_2_6, veo_3, veo_3_1, sora_2, seedance_lite, seedance_pro, seedance_pro_1_5, grok |
+| `video_from_text` | video | veo_3, veo_3_fast, veo_3_1, veo_3_1_fast, sora_2, kling_1_6, kling_2_1, kling_2_6, seedance-2.0, seedance-1.5-pro, seedance-1.0-pro, seedance-1.0-lite, grok |
+| `video_from_image` | video | kling_2_1, kling_2_6, veo_3, veo_3_1, sora_2, seedance-2.0, seedance_lite, seedance_pro, seedance_pro_1_5, grok |
 | `video_from_ingredients` | video | pika, kling_1_6, seedance_lite, veo_3_1, veo_3_1_fast, grok |
 | `speech_from_text` | audio | (voice_method: my_voices, design_voice, clone_voice) |
 | `lipsync` | video | sync_so, gen |

From dc8e2ad004de68b498f82e15a6e6bc59d48a786b Mon Sep 17 00:00:00 2001
From: mavneox <mav@onlywithneo.com>
Date: Thu, 30 Jul 2026 09:15:59 -0500
Subject: [PATCH 7/7] fix(docs): replace non-existent gen_upload_asset /
 social/upload_asset
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The tool gen_upload_asset and endpoint POST /v1/social/upload_asset do
not exist as public developer surfaces. Replace with the correct
three-step direct-upload flow: gen_create_direct_upload → PUT →
gen_complete_asset_upload, documented with curl recipe.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
---
 public/llms-full.txt | 22 ++++++++++++++++------
 public/llms.txt      |  2 +-
 2 files changed, 17 insertions(+), 7 deletions(-)

diff --git a/public/llms-full.txt b/public/llms-full.txt
index e726b7d..f6a5ac3 100644
--- a/public/llms-full.txt
+++ b/public/llms-full.txt
@@ -1108,17 +1108,27 @@ X thread:
 
 `thread` and `reply_to_tweet_id` are ignored for TikTok/Instagram/Facebook/YouTube. A thread that fails partway stops at the failed segment and is marked failed (not retried, to avoid duplicates).
 
-### Upload media (upload_asset)
+### Upload media (direct upload)
 
-Upload a local file to get a hosted CDN URL for use in `media_url`/`media_urls`.
+Upload a local file to get a hosted CDN URL for use in `media_url`/`media_urls`. Three-step flow:
 
 ```
-POST /v1/social/upload_asset   (multipart/form-data, api.gen.pro)
-Fields: file=<binary>, agent_id=<agent_id>
-→ { url, asset_id, content_resource_id, asset_registered }
+# Step 1: get a pre-signed upload URL
+POST /v1/direct_upload
+{ blob: { filename, byte_size, checksum, content_type } }
+→ { signed_id, direct_upload: { url, headers } }
+
+# Step 2: PUT the file to the signed URL (no auth header needed)
+PUT <direct_upload.url>  (file bytes, Content-Type header from direct_upload.headers)
+
+# Step 3: register the asset in GEN's library
+POST /v1/content_resources?agent_id={id}
+{ content_resource: { file: <signed_id> } }
+→ { id, url, ... }
 ```
 
-For up to 4 images on X, call `upload_asset` per file and pass all URLs in `media_urls`.
+MCP tools: `gen_create_direct_upload` (Step 1) → `gen_complete_asset_upload` (Step 3).
+For up to 4 images on X, upload each file and pass all resulting URLs in `media_urls`.
 
 ### Content calendar
 
diff --git a/public/llms.txt b/public/llms.txt
index 828facf..186b752 100644
--- a/public/llms.txt
+++ b/public/llms.txt
@@ -818,7 +818,7 @@ The current MCP exposes 142 `gen_*` tools, 9 prompts, and 3 guidance resources.
 
 **Creation:** `gen_create_video` (single clip; models: seedance-2.0, veo_3, kling_2_1, sora_2, grok, pika), `gen_create_image` (nano-banana-pro/nano-banana/nano-banana-2), `gen_create_song` (suno-v5-beta/suno-v4.5plus-beta), `gen_create_song_mix`, `gen_render_video`, `gen_generate_content` (trigger generation for a vidsheet cell), `gen_generate_layer` (trigger generation for a specific layer), `gen_continue_generation` (resume a stopped generation).
 
-**Assets / Content Resources:** `gen_list_assets` / `gen_list_content_resources`, `gen_get_asset` / `gen_get_content_resource`, `gen_import_asset_from_url`, `gen_upload_asset`, `gen_complete_asset_upload`, `gen_create_content_resource`, `gen_update_content_resource`, `gen_delete_asset` / `gen_delete_content_resource`.
+**Assets / Content Resources:** `gen_list_assets` / `gen_list_content_resources`, `gen_get_asset` / `gen_get_content_resource`, `gen_import_asset_from_url`, `gen_create_direct_upload` (get pre-signed S3 URL → PUT file → `gen_complete_asset_upload`), `gen_create_content_resource`, `gen_update_content_resource`, `gen_delete_asset` / `gen_delete_content_resource`.
 
 **Voices:** `gen_list_agent_voices`, `gen_list_my_voices`, `gen_design_voice`, `gen_clone_voice`, `gen_preview_voice` (TTS audition), `gen_get_voice_preview_status`, `gen_generate_voice_description`, `gen_generate_voice_samples`, `gen_get_voice_sample_status`, `gen_delete_voice`.