Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -26,6 +26,7 @@ Most editors keep a project in a format built for their own app, not for you or
- **Transitions**: put a dissolve, a fade through black or white, a wipe, a slide, a blur or a pixelate on any cut, from the mark between two clips on the timeline, or on every cut at once. A transition is centred on its cut and moves no clip, so titles and music stay where you put them, and the preview draws each one frame for frame as it exports.
- **Captions**: switch them on for the whole video, pick one of twelve languages and one style. The words come from each clip's transcript, so captions follow every trim, split and reorder.
- **AI, in one conversation**: add your clips and say what you want beside the cut, in your own words: it builds the cut from them ("a 30 second video from my clips") or changes the one you have. It remembers the conversation, so "shorter" or "a square one too" knows what it refers to, and it asks back instead of guessing when a request could mean two things. A request that would watch a lot of footage says how much first and, on a yes, runs in the background: keep editing or close the tab, and it never lands on top of a change you made meanwhile. The model calls a fixed set of checked operations, so the edit is always a valid document. One request, one undo.
- **Clips from a long video**: "5 vertical clips with a strong hook from this interview" reads what is said, picks the moments that stand on their own, opens each on its strongest line, and makes each its own 9:16 project with captions, framed on the person speaking. Shots it is unsure how to frame keep the whole picture, and it tells you which. "Frame it for vertical" does the same for the cut you have.
- **AI assist**: Auto-cut watches several clips together and assembles the strongest sequence for what the video is for, captions included; Clean up trims one clip down to its good parts.
- **Export to MP4** in draft, standard or high quality, on Clawnify's managed edit service.
- **Share by link**: one link per project plays an export to anyone you send it to, with no sign-in, and a Download button. The link stays on the export you shared until you move it to a newer one, so a draft never reaches viewers; turning it off ends it for everyone.
Expand Down
31 changes: 30 additions & 1 deletion agent.md
Original file line number Diff line number Diff line change
Expand Up @@ -217,6 +217,35 @@ video from my clips" works from an empty timeline: it places a video with
`add_clip`, or builds the whole main track from up to 8 of them with
`assemble` (Auto-cut's pass, which replaces the main track).

### Asked in the Clawnify chat

When a person asks you, in the dashboard's chat, to change a video project or
make videos from footage, send their words to `POST
/api/projects/{id}/instruct` rather than writing the document yourself: it
watches the footage, picks the steps (cutting, assembling, clips, framing) and
applies them as one step the person can undo, never over an edit they made
meanwhile. The editor beside the chat tells you which project is open (a
`project` record in your context). Send `background: true` when the request
watches footage (clips, framing, clean-up, "make a video from my clips"), then
tell the person it is running: the editor shows its progress. Read `GET
/api/projects/{id}/ask` for how it ended, and pass on what `said` says,
including any shot it was unsure how to frame.

### Clips from a long video, and framing for vertical

"5 vertical clips with a strong hook from this interview" is one request:
`/instruct` reads the video's transcript, picks the moments that stand on
their own (about 45 seconds unless asked otherwise, at most 8), opens each on
its strongest line (played first when it comes later in the moment), and makes
each one **its own project**: 1080×1920 with captions on, framed on the person
speaking. The project the request was sent to is not changed. The finished
message, and the job while it runs, list them in `made: [{ id, name }]`; open
one at `/edits/{id}`. "Frame it for vertical" frames the project's own cut the
same way, in its 9:16 version (added when missing). Shots the framing is unsure
of keep the whole picture, and the message says how many: a person checks them
in the editor's AI reframe. Both need the video on the media service (an
upload from the editor or a Drive import) and always run in the background.

`/instruct` is the same conversation people see in the editor's AI panel: it
reads the last few turns, so a follow-up ("shorter") works, and your requests
show there beside theirs. When a request could mean two things, or would watch
Expand Down Expand Up @@ -317,7 +346,7 @@ in line, and asks Drive again now for the clips waiting on it. `DELETE /api/proj
| PUT | `/api/projects/{id}` | Update `{ name?, brief?, edl? }` — the EDL is validated on save |
| POST | `/api/projects/{id}/autocut` | `{ asset_ids, prompt? }` — AI assembles the main track from several clips |
| POST | `/api/projects/{id}/instruct` | `{ instruction, background? }` — change the existing cut in words → `{ edl, said, applied, asked, job? }`; `asked: true` means `said` is a question back (or, with `job`, a proposal) and nothing was changed. `background: true` → `202` with a queued `job` |
| GET | `/api/projects/{id}/ask` | The project's AI conversation, oldest first, and its recent background jobs → `{ messages: [{ role, text, author, applied, asked, failed, job_id }], jobs }` |
| GET | `/api/projects/{id}/ask` | The project's AI conversation, oldest first, and its recent background jobs → `{ messages: [{ role, text, author, applied, asked, failed, job_id, made }], jobs }` (`made`: projects a request made) |
| GET | `/api/projects/{id}/ask/jobs/{job}` | One background job with `base_edl` (the cut it started from), `draft_edl` (its cut so far, while it runs) and `result_edl` |
| POST | `/api/projects/{id}/ask/jobs/{job}/run` | Yes to a proposal: run it in the background → `202` |
| POST | `/api/projects/{id}/ask/jobs/{job}/cancel` | Drop a proposal, stop a queued or running job (its result is not applied), or let go of a kept one |
Expand Down
4 changes: 2 additions & 2 deletions package.json
Original file line number Diff line number Diff line change
Expand Up @@ -12,9 +12,9 @@
"test:e2e": "esbuild src/server/index.ts --bundle --format=esm --platform=node --outfile=test/e2e/.server.mjs --log-level=warning && node --no-warnings test/e2e/run.mjs ./.server.mjs schema.sql && node --no-warnings test/e2e/drive-imports.mjs"
},
"dependencies": {
"@clawnify/app": "^0.2.1",
"@clawnify/app": "^0.4.5",
"@clawnify/connections": "^0.7.0",
"@clawnify/db": "^0.4.2",
"@clawnify/db": "^0.4.3",
"@clawnify/queue": "^0.1.1",
"@clawnify/services": "0.6.0",
"@fontsource-variable/inter": "^5.3.0",
Expand Down
Loading
Loading