Clone any viral video with AI agents. Not just a script, the whole workflow: swap the face, the words, the B-roll, ship 100 variants in one command, and get your 100M views.
TypeScript agentic-aiaiai-agentsai-videocompilerdslffmpeggenerative-aillmmarkup-languagemonorepoplugin-systemprogramming-languagetext-to-videotypescriptvideovideo-automationvideo-clonevideo-editingvideo-generation
docs: four rules the tier-list reconstruction paid for
rpone7cd540e1150d
diff --git a/.agents/skills/hypit/references/local-author-package.md b/.agents/skills/hypit/references/local-author-package.mdindex d1aeba75..e128e5fd 100644--- a/.agents/skills/hypit/references/local-author-package.md+++ b/.agents/skills/hypit/references/local-author-package.md@@ -114,6 +114,15 @@ nothing.`media-track:Item` and the `icon` port on `@hypit/ranking` are the shape this takes.- Never require an input the installing project has no reason to choose. If a Run Source exists onlyto produce the component's own texture, the texture is in the wrong place.+- **A slot that is not filled yet costs its own cell and nothing more.** The component's structure —+ its rows, bars, frames, labels, the geometry it computes — is drawn whatever the slots turn out to+ hold. Draw the elements whose material is there, leave the ones whose material is not, and measure+ every cell either way so nothing moves when the rest arrive. A component that refuses to draw+ anything because one slot holds material it cannot use has made the whole composition depend on+ its most incomplete input, and the symptom reads as a dead Track rather than as a missing picture:+ the Producer refuses, the Track never builds, and the rows nobody was waiting for vanish with it.+ This is what a Source looks like for the whole stretch between being written and being built, which+ is most of its life.- A component that cannot render on its own cannot produce the preview image its Surface owes. Treata missing preview as evidence of this mistake rather than a step to skip.diff --git a/.agents/skills/hypit/references/playbooks/craft/production-gates.md b/.agents/skills/hypit/references/playbooks/craft/production-gates.mdindex 4dcc8f62..73a6feff 100644--- a/.agents/skills/hypit/references/playbooks/craft/production-gates.md+++ b/.agents/skills/hypit/references/playbooks/craft/production-gates.md@@ -31,6 +31,12 @@ next. `generated-dependencies.md` says which images derive from which.accepted Record of each output, ready to paste. Every Record you fail to pin is a generation youpay for twice.+Pin as soon as a Build accepts something, not when you next want to build. **A Build that fails still+accepted everything upstream of the failure**, and that is exactly when pinning is skipped: the run+ended in an error, so it does not feel like a run that produced anything. It produced every picture+and every take that ran before the step that broke, and resubmitting without pinning them buys the+same bytes again. Run `--pin` against the failed Build before touching the cause.+Apply this checklist to every image:- identity: correct recurring person/product/location, complete face when required, no drift;diff --git a/.agents/skills/hypit/references/reference-video/final-sources.md b/.agents/skills/hypit/references/reference-video/final-sources.mdindex 5002f77d..e5fda4d9 100644--- a/.agents/skills/hypit/references/reference-video/final-sources.md+++ b/.agents/skills/hypit/references/reference-video/final-sources.md@@ -26,6 +26,21 @@ any Build is submitted, which is the correct place for the author to find out.Do not author one source fragment per shot. Do not let check success substitute for unresolvedsemantic evidence; return to a narrow `observe_reference` question when necessary.+## A take's duration is measured, not estimated++`estimate:Speech` predicts how long a line will take to say. In this route that prediction is the+wrong input, because the reference already contains the answer: `prepare_reference` measured every+word of it, so each Segment's real duration is the span from its first word's start to its last+word's end in `transcript_ref`. Write that number on the take.++An estimate is not close enough to skip this. A reconstruction whose Segments are each estimated a+second long finishes several seconds longer than the reference, and every reveal, cut and overlay+lands late against a program that no longer matches the thing it reconstructs. The estimate is for+original authoring, where nobody has said the words yet.++Keep the estimate only where the reference cannot answer: a line the reconstruction adds, or a+Segment whose speech the reference never contains.+Use existing checks only:```bashdiff --git a/.agents/skills/hypit/references/reference-video/workflow.md b/.agents/skills/hypit/references/reference-video/workflow.mdindex a0f32916..6bc660d6 100644--- a/.agents/skills/hypit/references/reference-video/workflow.md+++ b/.agents/skills/hypit/references/reference-video/workflow.md@@ -49,7 +49,8 @@ reconstructing a location depends on, since reference frames are never fed to geword, measured locally by WhisperX. It is not an observation: no model wrote it, nothing about it issent to Gemini, and it does not go in the observation cache. Read it whenever a decision depends onwhen a word is said — placing each on-screen text reveal against the line that triggers it, timing a-caption, or checking that a voice observation matches what was actually spoken. Do not run WhisperX+caption, checking that a voice observation matches what was actually spoken, or setting how long a+take runs, which `final-sources.md` measures from these words rather than estimating. Do not run WhisperXby hand and do not ask a model to transcribe: the transcript is already there.`transcript` reports `status`, `transcript_ref` and `word_count`. Read the words fromdiff --git a/.agents/skills/hypit/references/runtime.md b/.agents/skills/hypit/references/runtime.mdindex dd882249..56d111e6 100644--- a/.agents/skills/hypit/references/runtime.md+++ b/.agents/skills/hypit/references/runtime.md@@ -44,6 +44,18 @@ Builds remain archived and neither command cancels remote Provider work.- There is no implicit cache or Pin state. Reuse Records through a new `.svrun` containing`build-record` and `satisfy`; the Candidate supplies the exact nominal Type required by thecurrent Logical Output.+- **A running Worker holds the package code it loaded.** Edit a package — yours or an installed one —+ and the Worker that is already up keeps executing the version it started with, so the next Build+ fails exactly as the last one did. Stop it before resubmitting; the next `build` starts a fresh one+ that loads the edit:++ ```bash+ ps -eo pid,command | grep "hypit.mjs _worker" | grep <project>/hypit.runtime.json+ kill <pid>+ ```++ Read an identical repeat failure as this until you have ruled it out. Reasoning about why a correct+ fix did not work is how an afternoon goes, and the fix was never loaded.## Keep project and package boundaries distinct