The Video Generator
The Video Generator is the moving-picture sibling of the Image Generator: describe a shot, get a short clip. Product teasers, animated backgrounds, little social moments — generated on whatever video model your device or provider offers, delivered into the project's artifacts.
Like its sibling, it's fixed-function by design: no file editing, no workspace access, just generation. Short clips take real compute — on local hardware expect renders to be measured in minutes, not seconds, and lean on the smallest length that sells the idea.
Their character
What this gezel does
This gezel is a direct line to video generation. There's no chatbot in the middle — when you @mention it with a prompt, your text goes straight to the video engine on your machine and the finished clip lands in chat.
How to use it
Type the prompt as a plain description of the clip you want:
@{name} a paper boat sailing down a rain gutter, golden hour, slow motion
The clip is saved into your project's artifacts/generated/ folder, and a poster frame shows inline in the chat bubble. To iterate, ask again with a new prompt — it doesn't remember earlier messages, since there's no chatbot on this side.
To animate a starting image, attach an image to your message (on models that support it, like Wan 2.2 TI2V-5B).
Settings
The video model is picked in Settings → Video generation. Per-gezel defaults — width, height, frame count, frame rate, and model — are editable in the gezel's edit dialog under "Default arguments". Note that video generation is slow and uses the GPU exclusively, pausing the chat model while it runs.
What they can do
| Tool group | Purpose | Tools |
|---|---|---|
| Video Tools | Generate short video clips via the configured local video model (LTX / WAN) | generate_video |
| Task Management | Create, assign, advance, and report on tasks | list_tasks, get_task, list_craftbooks, suggest_craftbook, import_skill, create_task, start_plan, invoke_craftbook, update_task, set_outcomes, verify_outcome, add_verification_step, set_task_status, activate_task, list_task_children, add_task_step, advance_task_step, assign_task, read_task_notes, write_task_note, spawn_task_instances |
| Project Artifacts | Project-scoped read-write outputs (reports, scratch files, scripts a gezel produces, and large outputs auto-saved by tools that exceed the inline cap) | list_artifacts, read_artifact, write_artifact, grep_artifact |
| Shared document library | Cross-project shared library — mission docs, guidelines, and any markdown the user wants every gezel to be able to find | list_documents, read_document, write_document, delete_document, search_documents |
| Memory | Persistent notes a gezel can search, recall, and write back | search_memory, save_memory, list_memories |
| User Interaction | Pose a structured question mid-turn — to the user (ask_user_question), to a specific gezel (ask_gezel), or to a role-shaped specialist (ask_specialist) — instead of guessing | ask_user_question, ask_gezel, ask_specialist |
| Handboek | Consult gezel's built-in documentation for meta questions about gezel itself — roles, craftbooks, projects, memory, models, setup | how_do_i |
On this device
| Tier | Model size | Tool surface for this role |
|---|---|---|
| tiny | under 5B | full kit (38 tools) |
| small | 5–12B | full kit (38 tools) |
| medium | 12–45B | full kit (38 tools) |
| large | 45B and up | full kit (38 tools) |
| cloud | hosted | full kit (38 tools) |
As models grow
| Tier | Model size | Tool surface for this role |
|---|---|---|
| tiny | under 5B | full kit (38 tools) |
| small | 5–12B | full kit (38 tools) |
| medium | 12–45B | full kit (38 tools) |
| large | 45B and up | full kit (38 tools) |
| cloud | hosted | full kit (38 tools) |