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: two loops, two attempts each, and read the observation before the package
rpone8eb70b638271
diff --git a/.agents/skills/hypit/references/local-author-package.md b/.agents/skills/hypit/references/local-author-package.mdindex 0658f694..88196a00 100644--- a/.agents/skills/hypit/references/local-author-package.md+++ b/.agents/skills/hypit/references/local-author-package.md@@ -4,6 +4,21 @@ Use this workflow only after proving that no legal composition of installed packrequired behavior. A similar-looking tag is insufficient when its declared Types, timing or outputbehavior differs. Never write an unknown tag before its package exists.+## Read the observation before deciding the package's shape++The observation is the contract, and it is read before the first line of the package — not assumed,+not remembered, not rediscovered halfway. In the reference-video route the evidence for the element+this package owns lives in the shot observations (`type:` and `visual:` for its text and picture,+`persistent_systems` for its whole-reference life) and the word-level transcript. Read them.++Then write out, one line each, every appearance property the observation states — the typeface+character, the weights, the mix of faces, the colours, the stroke, the shadow, the motion, the+timing. A package whose declared vocabulary cannot express one of those lines is wrong before it is+written: either widen the package to carry it, or record it as an accepted deviation with its reason+**before** you start. The one thing that may not happen is authoring a package that silently drops a+stated property — that is how a reference whose title runs two typefaces ends up as a package that+draws one font, and the loop cannot repair what the package cannot express.+## Write `package.json` and the activation firstWrite those two before any implementation, run `pnpm install` once, and only then start on thediff --git a/.agents/skills/hypit/references/reference-video/reconstruction-loop.md b/.agents/skills/hypit/references/reference-video/reconstruction-loop.mdindex a1c3d347..d51e3731 100644--- a/.agents/skills/hypit/references/reference-video/reconstruction-loop.md+++ b/.agents/skills/hypit/references/reference-video/reconstruction-loop.md@@ -72,16 +72,31 @@ return a difference every round for ever. So the loop ends on whichever of these- **No progress ends it immediately.** If a comparison returns the same difference it returned beforethe repair, stop. The repair is not reaching the problem, and two more rounds of the same reasoningwill not find it. Rendering and comparing cost real time on every round.-- **Two aimed attempts per element is the ceiling.** The first fixes what is obvious, the second- fixes what the first revealed. A third round is where guessing starts to overshoot — correcting- past the reference rather than towards it — and the run has more elements waiting than any one of- them is worth. Time beats fidelity here by explicit choice.+- **There are two loops, each with its own two-attempt ceiling.** The first loop is over the+ *package*: render what it draws and compare it; a difference that the package cannot express is a+ package defect, and fixing it means changing the package's structure. The second loop is over the+ *values*: once the package can express everything the observation states, tune the Recipe, the+ font, the placement and the motion until it looks like the reference. The two never share attempts.++ **Write-and-remake the package: two attempts.** The first fixes what is obvious, the second fixes+ what the first revealed. If after two repairs the difference is still one the package cannot+ express — a typeface mix it has no port for, a motion it cannot draw — the package is wrong, not+ the values, and continuing to tune values is thrashing. Widen the package instead, which starts a+ fresh write-and-remake loop for the widened shape.++ **Fill the parameters: two attempts.** A difference the package *can* express — the wrong weight,+ the wrong size, the wrong colour, the wrong position — is a value, and tuning it is the second+ loop. Two attempts, then stop.++ Beyond each ceiling, guesses start to overshoot — correcting past the reference rather than+ towards it. Time beats fidelity here by explicit choice.## Measure what can be measured; iterate only on what cannot-Two attempts is not enough to converge by guessing, and it is not meant to be. A difference stated as-a quantity — a stroke that is too thick, a shape that is too tall, type that is too large, a margin-that is too wide — is not a guessing problem. Read the value off the reference frame directly:+Two attempts per loop is not enough to converge by guessing, and it is not meant to be. A difference+stated as a quantity — a stroke that is too thick, a shape that is too tall, type that is too large,+a margin that is too wide — is not a guessing problem. Read the value off the reference frame+directly:the frames are ordinary images in `.hypit/reference-video-tools/<reference-id>/shots/`, and measuringone against the known frame size gives the number in a single step.