The FortressLog in
← Back to projects

What you can do here

Ask your agent for any of this in plain English. No API knowledge needed.

Build it as a script or an app

Every workflow can be a script you run from the terminal, a small web app with its own page, or both. Say which you'd prefer.

Build Brian slideshows

Create slide templates, generate the images with AI, edit backgrounds, and group slides into clumps. Works from a script or in the UI.

Generate images, video, and audio

Generate and edit images, make AI video (including Seedance), and produce speech, music, dubbing, and sound effects.

Borrow from TikTok

Scrape slides, videos, music — even authors and whole accounts. Clone a slideshow post or an account. Build a sound group from an account's music.

Pick where a post goes

Send each post to an account group, specific accounts, or a random few of them. You pick the client and accounts once, up front.

Track results with your own template

Attach a tracking template to your posts so you can see how they perform, separate from everything else on the accounts.

Search the web

Search the web (via Brave) and use the results in a workflow — research a topic, find references, or check what's current.

See your queue and results

See what's staged, what actually published, and how the posts performed.

These are the highlights. The full list of services is below.

All services

Generated from the backend catalog, so it stays current. Open one to see what it does.

Account groups7 actions
  • List all account groups for a client.
  • Create or upsert an account group.
  • Update an account group's title, client ID, and flag settings.
  • Retrieve auto-assign activity logs for an account group.
  • Add accounts to an account group.
  • Assign templates to an account group.
  • Get active auto-assign template assignments for one or more account g…
Accounts6 actions
  • List all accounts for a client.
  • Update settings for an account.
  • Assign templates to an ungrouped account.
  • Update video and image settings for an account.
  • Get recent posts for one or more accounts.
  • Get video and image settings for all accounts in a client.
Assets4 actions
  • List assets you can see.
  • Delete an asset by id.
  • Get a single asset by id.
  • Upload an asset (image or video) and get a hosted URL back.
Audio8 actions
  • Generate multi-speaker dialogue audio with ElevenLabs (async).
  • Dub an audio/video file into another language with ElevenLabs (async).
  • Generate music from a text prompt with ElevenLabs (async).
  • Generate a sound effect from a text prompt with ElevenLabs (async).
  • Generate speech audio from text with ElevenLabs (text-to-speech, asyn…
  • Transcribe an audio file to text (OpenAI Whisper or ElevenLabs Scribe…
  • Transform speech in an audio file into a target voice with ElevenLabs…
  • List/search available ElevenLabs voices.
Clients1 action
  • List and search active clients.
Creative groups10 actions
  • List and search creative groups.
  • Create an image or prompt creative group and populate its content.
  • Get a creative group with its content items.
  • Update an image or prompt creative group's name and content items.
  • Fetch performance metrics for a creative group.
  • Fetch performance metrics for individual items within a creative grou…
  • Create a copy group (captions or textbox text) and populate its lines.
  • Update a copy group's name and text lines.
  • Create a sound group from TikTok music URLs.
  • Update a sound group's name and TikTok music URLs.
Images8 actions
  • Analyze image(s) with a vision model — describe, extract text, or ans…
  • Edit or compose images with GPT Image from a prompt and source image…
  • Edit or compose images with the Nano Banana model from a prompt and s…
  • Generate an image with GPT Image from a text prompt.
  • Generate several related images as a grid with the GPT Image model, t…
  • Generate an image with the Nano Banana model from a text prompt.
  • Generate several related images as a grid with the Nano Banana model,…
  • Remove an image's background via MuAPI (transparent cutout).
Identity1 action
  • Verify your internal API key and return your identity.
Posts5 actions
  • List staged posts (the publishing queue), filtered and paginated.
  • Stage a post (video or slideshow) to an account's posting queue.
  • Get a staged post by its internal_post_id, including its lifecycle st…
  • Cancel (discard) a staged post that has not started publishing yet.
  • Per-post daily incremental views/engagement derived from post_metrics…
Prompts4 actions
  • Generate one edit instruction per image for use in a downstream image…
  • Generate a prompt from freeform instructions.
  • Generate a single text-to-image prompt from a text description.
  • Generate one text-to-image prompt per image by visually analyzing eac…
Published posts1 action
  • List published posts with their performance metrics, filtered and pag…
Runs1 action
  • Poll the status and result of an async operation by its runId.
Templates7 actions
  • List and search templates across all content types.
  • Create a new template.
  • Get one template of any content type.
  • List and search Brian slide templates.
  • Get one Brian slide template with its slides and clumps.
  • Update a Brian slide template.
  • Generate slide preview images for a Brian slide template.
TikTok4 actions
  • Clone slideshow posts from a TikTok account, optionally creating one…
  • Clone a single TikTok post (slideshow), optionally creating a templat…
  • Flexible TikTok scrape (profiles, hashtags, search, or post URLs) — t…
  • Scan a TikTok account's sounds, optionally building a deduplicated so…
Videos26 actions
  • Analyze a video with AI — describe, summarize, or answer a prompt abo…
  • Composite full-frame backgrounds (image or video) with optional pictu…
  • Concatenate multiple videos end-to-end into a single video.
  • Render a standalone CTA/plug clip from typed fields (no TSX).
  • Extract a video's audio track to an mp3.
  • Generate a Remotion composition (TSX) from a natural-language prompt…
  • Animate a first-frame image into a video via MuAPI (Kling, Veo 3.1, S…
  • Lip-sync a face to an audio track via MuAPI (talking-head / avatar vi…
  • Generate a video from reference images via MuAPI omni/multi-reference…
  • Generate a video from a text prompt via MuAPI (Veo, Sora, Kling, Seed…
  • Remove a video's audio track.
  • Edit a video (URL or previous Omni result) with Gemini Omni Flash.
  • Animate a first-frame image into a video with Gemini Omni Flash.
  • Generate a video from style/character/object reference images with Ge…
  • Generate a video from a text prompt with Gemini Omni Flash.
  • Read the durations of one or more videos.
  • Render an arbitrary Remotion composition (TSX source + assets) — the…
  • Downscale a video so its shorter side is at most maxShortSidePx.
  • Animate a first-frame image into a video with Seedance 2.0 via Atlas…
  • Generate a video from reference images/videos with Seedance 2.0 via A…
  • Generate a video from a text prompt with Seedance 2.0 via Atlas Cloud.
  • Speed a video up or down.
  • Grab a single frame from a video as an image.
  • Trim a video to a time range (optionally mute and/or downscale).
  • Analyze a previously-indexed video and generate text about it with Tw…
  • Index a video into TwelveLabs so it can later be analyzed.
Web2 actions
  • Fetch a web page and extract its readable text.
  • Search the web via Brave and return trimmed results.

What's new

2026-08-07

  • Contributors without a GitHub org seat can now work through the shared webmaster account. Project ownership and generated-by attribution come from each member's user-bound Fortress API key (GET /me), while existing contributors can keep using their personal GitHub account.

2026-07-30

  • Added the mistplay-travel-teasers project: automated travel-teaser videos (realistic AI still → Omni animation, 15 clips, text hook + Mistplay CTA) that post on a schedule across a Mistplay account group, never repeating a destination on the same account.

2026-07-21

  • The slide editor app that Nikola created is now available to use.

2026-07-17

  • You can now generate grid of images and can extract them. This is useful for generating assets cheaply, and use them to generate videos using remotion.

2026-07-10

  • You can now search the web (via Brave) and use the results in your workflows.
  • You can now use Google OMNI - text-to-video, image-to-video, reference-to-video, edit

2026-07-09

  • Added npm run update to pull the latest changes and see what's new.
  • npm run doctor now warns you when your copy of the repo is behind.
  • Added a "What you can do" page to the UI.
  • The agent now asks which client, which accounts or account group, and whether you want a tracking template before writing a posting script.

2026-07-07

  • Added the Brian slideshow toolkit: slide templates, AI images, AI-edited backgrounds, and clumps.

2026-07-04

  • Code owners can now wire a direct provider integration inside their own project as a temporary bridge.

2026-06-24

  • Asset uploads now support big files.