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: point the rules at the commands that now exist

rpone73a5f3e5b38f
diff --git a/.agents/skills/hypit/references/local-author-package.md b/.agents/skills/hypit/references/local-author-package.md
index bd4db621..58ebd4ff 100644
--- a/.agents/skills/hypit/references/local-author-package.md
+++ b/.agents/skills/hypit/references/local-author-package.md
@@ -67,11 +67,16 @@ nothing.
backdrop — belongs to the package. Commit it as an ordinary file under the package and read it
with `readFile(new URL("../assets/…", import.meta.url))`, the same way the Surface preview image
is read. Generate it once while authoring the package if it does not exist.
-- Produce that file outside the graph. Call the image provider's HTTP API directly with the
- credentials already in the environment, or use any equivalent tool, and save the bytes into the
- package. A package asset is authoring input, not the output of anybody's video: it is never a
- Target, never a Record, and never a reason to write a Run Source. Writing one to obtain it is the
- mistake this section exists to prevent.
+- Produce that file with `hypit image`:
+
+ ```bash
+ hypit image --prompt "the surface this component draws on" --to packages/local-<slug>/assets/paper.png
+ ```
+
+ It writes a picture and nothing else — no Source, no Build, no Record, no Runtime Profile. A
+ package asset is authoring input, not the output of anybody's video, so it is never a Target and
+ never a reason to write a Run Source. Writing one to obtain it is the mistake this section exists
+ to prevent.
- Only pictures that differ between videos are input ports: photographs, screenshots, thumbnails,
product shots, character images.
- Never require an input the installing project has no reason to choose. If a Run Source exists only
diff --git a/.agents/skills/hypit/references/playbooks/craft/graphic-compositions.md b/.agents/skills/hypit/references/playbooks/craft/graphic-compositions.md
index 2291bf01..0d9fb401 100644
--- a/.agents/skills/hypit/references/playbooks/craft/graphic-compositions.md
+++ b/.agents/skills/hypit/references/playbooks/craft/graphic-compositions.md
@@ -70,9 +70,9 @@ project that installs it to obtain a picture that was never theirs to choose. If
writing a Run Source whose only purpose is to produce a component's own texture, the texture is in
the wrong place: generate it once while authoring the package and commit the file.
-Produce that file outside the graph — call the image provider's HTTP API directly, or use any
-equivalent tool, and save the bytes into the package. A package asset is authoring input, not the
-output of anybody's video, so it is never a Target and never a Record.
+Produce that file with `hypit image --prompt <text> --to <path>`, which writes a picture and nothing
+else. A package asset is authoring input, not the output of anybody's video, so it is never a Target
+and never a Record.
A package is installed vocabulary. `.svml`, `.svs` and `.svrun` are documents that use it. A package
that needs one of those documents in order to draw itself has inverted that relationship.
diff --git a/.agents/skills/hypit/references/playbooks/craft/production-gates.md b/.agents/skills/hypit/references/playbooks/craft/production-gates.md
index f9f56bf6..733b308e 100644
--- a/.agents/skills/hypit/references/playbooks/craft/production-gates.md
+++ b/.agents/skills/hypit/references/playbooks/craft/production-gates.md
@@ -20,7 +20,10 @@ Use staged Run Sources so unreviewed expensive outputs cannot silently feed late
4. Reject and regenerate any failed image. Do not connect a failed image to Seedance, Media Track,
image composition, or another downstream generator.
5. Reuse accepted image Records with `.svrun` `build-record` and connect them to the next Build with
- `satisfy`. Changing the Target list alone does not select those Records.
+ `satisfy`. Changing the Target list alone does not select those Records. Do not transcribe the
+ Build ids by hand — `hypit history --source <author.svml> --pin` emits the pairs for the newest
+ accepted Record of each output, ready to paste. Every Record you fail to pin is a generation you
+ pay for twice.
Apply this checklist to every image:
diff --git a/.agents/skills/hypit/references/reference-video/vocabulary.md b/.agents/skills/hypit/references/reference-video/vocabulary.md
index fdeb2018..a45592a1 100644
--- a/.agents/skills/hypit/references/reference-video/vocabulary.md
+++ b/.agents/skills/hypit/references/reference-video/vocabulary.md
@@ -7,8 +7,16 @@ attributes, children, ports, Recipe properties, timing behavior, appearance and
First enumerate every system the reference actually contains — base pictures, inserted elements,
full-screen graphic compositions, persistent overlays, captions, speech, music, sound effects — and
inspect candidates for each one. A system you never inspected is a system you are about to invent.
-`inspect_svml_vocabulary` does not list what is installed; it reads the packages you name. Discover
-candidate names from `docs/guide/packages.md` and the `packages/` directory before inspecting.
+`inspect_svml_vocabulary` does not list what is installed; it reads the packages you name. Get the
+names from `list_svml_packages`, which reports every installed package that declares an activation
+along with the tags it registers:
+
+```bash
+hypit-reference-video-tools list_svml_packages
+```
+
+Never work from a remembered list of package names, and never conclude a capability is missing
+because you did not think of the package that has it.
Use this decision order:
diff --git a/.agents/skills/hypit/references/reference-video/workflow.md b/.agents/skills/hypit/references/reference-video/workflow.md
index 2ef4a67d..02767b72 100644
--- a/.agents/skills/hypit/references/reference-video/workflow.md
+++ b/.agents/skills/hypit/references/reference-video/workflow.md
@@ -1,8 +1,9 @@
# Reference-video workflow
-The public interface is exactly four CLI subcommands from `@hypit/reference-video-tools`:
+The public interface is exactly five CLI subcommands from `@hypit/reference-video-tools`:
```bash
+hypit-reference-video-tools list_svml_packages
hypit-reference-video-tools prepare_reference --video-path <path>
hypit-reference-video-tools observe_reference --reference-id <reference-id>
hypit-reference-video-tools inspect_svml_vocabulary --package <package> --tag <tag>
@@ -12,7 +13,8 @@ hypit-reference-video-tools compare_reconstruction --reference-id <reference-id>
Defaults are sufficient for normal use. Each also accepts `--input <json>`. Follow this sequence:
```text
-prepare_reference
+list_svml_packages
+→ prepare_reference
→ observe_reference for all shots
→ narrow observe_reference questions for unresolved appearance and conflicts
→ inspect_svml_vocabulary for candidate packages
@@ -22,6 +24,13 @@ prepare_reference
→ render each authored element and close the loop with compare_reconstruction
```
+## list_svml_packages
+
+Reports every installed package that declares an activation, with the Surface tags it registers.
+This is the only way to know what vocabulary exists: `inspect_svml_vocabulary` reads packages you
+name, and the Build CLI is deliberately unable to scan a directory. Run it before deciding anything
+is missing.
+
## prepare_reference
`prepare_reference` deterministically prepares clips, representative frames, tail frames, audio tails
diff --git a/packages/reference-video-tools/README.md b/packages/reference-video-tools/README.md
index 950df8ae..d6290816 100644
--- a/packages/reference-video-tools/README.md
+++ b/packages/reference-video-tools/README.md
@@ -2,8 +2,8 @@
CLI tools for reconstructing a reference video with Hypit.
-The package exposes four CLI subcommands: `prepare_reference`, `observe_reference`,
-`inspect_svml_vocabulary`, and `compare_reconstruction`. Each command prints one JSON result to
+The package exposes five CLI subcommands: `list_svml_packages`, `prepare_reference`,
+`observe_reference`, `inspect_svml_vocabulary`, and `compare_reconstruction`. Each command prints one JSON result to
stdout. Every command except `inspect_svml_vocabulary` sends narrow natural-language requests to
Gemini with a fixed temperature of `1.0`; they never receive SVML syntax and never write SVML. The
final source files are authored by the calling agent and checked with the existing
@@ -12,6 +12,7 @@ final source files are authored by the calling agent and checked with the existi
Run the CLI from the repository or an installed package:
```bash
+pnpm hypit-reference-video-tools list_svml_packages
pnpm hypit-reference-video-tools prepare_reference --video-path ./reference.mp4
pnpm hypit-reference-video-tools prepare_reference --video-path ./reference.mp4 --redo people
pnpm hypit-reference-video-tools observe_reference --reference-id <reference-id>
@@ -25,6 +26,11 @@ pnpm hypit-reference-video-tools compare_reconstruction --reference-id <referenc
For automation, every command also accepts `--input '{"...":"..."}'` with the complete JSON input
object. This is useful when the calling agent already has a structured request.
+`list_svml_packages` reports every installed package that declares an activation, with the Surface
+tags it registers. `inspect_svml_vocabulary` reads packages you can already name, and the Build CLI
+is deliberately unable to scan a directory, so this is how a caller finds out what vocabulary exists
+instead of working from memory.
+
`prepare_reference` requires a local video path and stores derived media under the project's
gitignored `.hypit/reference-video-tools/` directory. It produces three full-reference observations:
`people_and_product`, `voices`, and `persistent_systems` — the on-screen text and graphic systems