Skip to content

armgit / hypit

Public Imported from GitHub
0

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.

docs: the reference is seen by Gemini, never by the agent

rpone245a8b74176c
diff --git a/.agents/skills/hypit/references/reference-video/reconstruction-loop.md b/.agents/skills/hypit/references/reference-video/reconstruction-loop.md
index c77e0234..0304b3dc 100644
--- a/.agents/skills/hypit/references/reference-video/reconstruction-loop.md
+++ b/.agents/skills/hypit/references/reference-video/reconstruction-loop.md
@@ -37,6 +37,22 @@ card:
`.hypit/reference-video-tools/<reference-id>/shots/NNN-representative.jpg`.
3. Repair, then render again.
+### The reference is seen by Gemini, never by you
+
+The reference frame is looked at exactly one way: through `compare_reconstruction`, which sends it
+and the rendered image to Gemini as an unlabelled pair and returns the differences in words. You do
+not open the reference frame yourself and look at it. Whatever visual ability you have — the model
+running this loop may be able to read images directly — is not to be used on the reference or on the
+rendered reconstruction. There is one observer of the reference, the same VLM that wrote the
+observations, and it reports through `compare_reconstruction`. A second observer is a second opinion
+that is paid for with the very bias it claims to correct: it sees the reconstruction, knows what was
+built, and confirms what it expects.
+
+This is why a font that is wrong is caught: `compare_reconstruction` names it, and the package is
+repaired. It is not caught by looking at the frame yourself and choosing a font that happens to
+resemble it — that route silently depends on the loop model having vision, and stops working the
+moment it does not.
+
The SVML Playground in `../preview.md` is a browser preview for a person to look at. It is not a
source of the image this loop needs.
@@ -100,10 +116,10 @@ return a difference every round for ever. So the loop ends on whichever of these
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 measuring
-one against the known frame size gives the number in a single step.
+a margin that is too wide — is not a guessing problem. Ask Gemini for the number: a
+`compare_reconstruction --question "how tall is the oval relative to the frame?"` over the shot that
+shows it most clearly returns a measurement in one step. The reference is seen by Gemini, never by
+you, and this is how a quantity is measured without ever opening the frame yourself.
Do that instead of spending an attempt. An attempt is for differences that have no number — a
typeface's character, a texture, a rhythm — where the only route is change it and look again.
diff --git a/.agents/skills/hypit/references/reference-video/workflow.md b/.agents/skills/hypit/references/reference-video/workflow.md
index 127ecab1..a0f32916 100644
--- a/.agents/skills/hypit/references/reference-video/workflow.md
+++ b/.agents/skills/hypit/references/reference-video/workflow.md
@@ -108,15 +108,18 @@ reported letters being *deleted* from the screen where the text was only typing
The frames are already on disk at
`.hypit/reference-video-tools/<reference-id>/shots/NNN-representative.jpg`, and the shot clips beside
-them. Opening one costs nothing and settles the question outright, so when evidence disagrees —
-one shot against the next, a shot against a whole-reference pass, or an observation against what the
-video plainly is — look at the frame and decide from it. Do not re-observe: that is paid, and at
-temperature `1.0` it returns a paraphrase rather than a correction.
+them. When evidence disagrees — one shot against the next, a shot against a whole-reference pass, or
+an observation against what the video plainly is — settle it by asking Gemini to look, through a
+narrow `observe_reference --question` over that shot, not by opening the frame yourself. You do not
+view the reference; `reconstruction-loop.md` says this is a rule, not a preference. Do not re-observe
+the full shot: that is paid, and at temperature `1.0` it returns a paraphrase rather than a
+correction.
Be most suspicious of anything an observation asserts about change over time — something appearing,
vanishing, being removed, being drawn in a single frame. A describer working from one pass infers
-those rather than seeing them, and infers them wrongly. Extract a few frames across the stretch with
-`ffmpeg` and look at the sequence yourself before building anything on such a claim.
+those rather than seeing them, and infers them wrongly. A `compare_reconstruction` against a rendered
+probe, or a narrow question over the stretch, is how such a claim is checked — not by extracting
+frames and reading the sequence yourself.
## Narrow questions
@@ -138,6 +141,11 @@ and returns a description of their visible differences. It is never told which i
was built, or how; `--question` may narrow it to one region of the picture and nothing else. Results
are not cached. `reconstruction-loop.md` governs when and how to use it.
+This is the only way the reference is ever seen. Do not open the reference frame yourself and look
+at it, even if the model running this route can read images — the reference has one observer, the
+same Gemini that wrote the observations, and it reports through this command. `reconstruction-loop.md`
+states this as a rule, not a preference.
+
## Boundaries
Gemini returns natural language only. Never send it SVML syntax, package declarations, vocabulary,