diff --git a/docs/guides/rendering.mdx b/docs/guides/rendering.mdx
index d81cb7e2392..30f59985b8f 100644
--- a/docs/guides/rendering.mdx
+++ b/docs/guides/rendering.mdx
@@ -130,7 +130,7 @@ browser already supports the source or you are diagnosing the proxy itself.
 
 ## Choose quality and frame rate
 
-The default `standard` quality is the right choice for most finished work.
+The default `looks` quality (CRF 16) is the right choice for most finished work.
 MOV is an alpha-preserving editing intermediate and always uses the fixed
 ProRes 4444 profile. `--crf` and `--video-bitrate` do not map to that profile
 and are rejected for MOV; choose MP4 or WebM when you need those controls.
diff --git a/docs/packages/cli.mdx b/docs/packages/cli.mdx
index e325024d42c..1cafed0ce86 100644
--- a/docs/packages/cli.mdx
+++ b/docs/packages/cli.mdx
@@ -306,7 +306,7 @@ npx hyperframes transcribe video.mp4 --to vtt
 | ----------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
 | `--dir, -d`       | Project directory (default: current directory)                                                                                                                                                     |
 | `--engine, -e`    | ASR engine: `auto` (default — Parakeet if installed and the language is one of its 25, else whisper), `parakeet` (fails if Parakeet is missing, broken, or lacks the `--language`), or `whisper`. Under `auto` or `parakeet`, a `--language` other than English runs whisper when it is installed, with a note, because Parakeet detects the language itself |
-| `--model, -m`     | Whisper model (default `small.en`): `tiny.en`, `base.en`, `small.en`, `medium.en`, `large-v3`                                                                                                       |
+| `--model, -m`     | Whisper model (default `small.en`): `tiny.en`, `base.en`, `small.en`, `medium.en`, `small`, `medium`, `large-v3`                                                                                                       |
 | `--language, -l`  | Language code (`en`, `es`, `ja`, …). Whisper transcribes as this language; Parakeet detects the language itself, so it only takes a non-English `--language` when whisper is not installed. |
 | `--to`            | Write an `srt` or `vtt` sidecar: from a transcript file it exports instead of importing; from audio or video it is written beside `transcript.json`, with cues for languages without spaces grouped, which exporting that `transcript.json` later may split differently |
 | `--output, -o`    | Where the exported SRT/VTT goes                                                                                                                                                                    |
@@ -1434,7 +1434,7 @@ The flags that come up most:
 
 | Flag                | Default                       | Description                                                                                                                                       |
 | ------------------- | ----------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------- |
-| `--output, -o`      | `renders/<name>.mp4`          | Output file path                                                                                                                                  |
+| `--output, -o`      | `renders/<project>_<YYYY-MM-DD>_<HH-MM-SS>.<ext>` | Output file path                                                                                                                                  |
 | `--composition, -c` | `index.html`                  | Render a different composition file. Sub-compositions using `<template>` wrappers must be referenced from `index.html` via `data-composition-src`. |
 | `--format`          | `mp4`                         | `mp4`, `webm`, `mov`, `gif`, `png-sequence`, or `hls`. WebM and MOV carry transparency; `png-sequence` writes RGBA frames to a directory; `hls` writes a VOD playlist directory. |
 | `--fps, -f`         | root `data-fps`, otherwise 30 | 1–240, or an ffmpeg rational like `30000/1001` for 29.97                                                                                            |
diff --git a/packages/cli/src/commands/render.ts b/packages/cli/src/commands/render.ts
index 30e3fcbdfab..1e2d0be9d57 100644
--- a/packages/cli/src/commands/render.ts
+++ b/packages/cli/src/commands/render.ts
@@ -159,7 +159,7 @@ export default defineCommand({
     output: {
       type: "string",
       alias: "o",
-      description: "Output path (default: renders/<name>.mp4)",
+      description: "Output path (default: renders/<project>_<YYYY-MM-DD>_<HH-MM-SS>.<ext>)",
     },
     fps: {
       type: "string",
diff --git a/packages/cli/src/commands/transcribe.ts b/packages/cli/src/commands/transcribe.ts
index bc9bec032c4..d0f14f60b9f 100644
--- a/packages/cli/src/commands/transcribe.ts
+++ b/packages/cli/src/commands/transcribe.ts
@@ -56,7 +56,7 @@ export default defineCommand({
     },
     model: {
       type: "string",
-      description: `Whisper model (default: ${DEFAULT_MODEL}). Options: tiny.en, base.en, small.en, medium.en, large-v3`,
+      description: `Whisper model (default: ${DEFAULT_MODEL}). Options: tiny.en, base.en, small.en, medium.en, small, medium, large-v3`,
       alias: "m",
     },
     language: {
diff --git a/packages/cli/src/docs/rendering.md b/packages/cli/src/docs/rendering.md
index 86480c80447..f7963df7a86 100644
--- a/packages/cli/src/docs/rendering.md
+++ b/packages/cli/src/docs/rendering.md
@@ -14,8 +14,8 @@ Requires: Docker installed and running.
 
 ## Options
 
-- `-f, --fps` — 24, 30, or 60 (default: 30)
-- `-q, --quality` — draft, standard, high (default: standard)
+- `-f, --fps` — 1-240 or a rational such as 30000/1001 (default: root `data-fps`, else 30)
+- `-q, --quality` — draft, looks, delivery, or standard/high (default: looks)
 - `-w, --workers` — Parallel workers 1-8 (default: auto)
 - `--crf` — Override encoder CRF (mutually exclusive with `--video-bitrate`)
 - `--video-bitrate` — Target video bitrate such as `10M` (mutually exclusive with `--crf`)
diff --git a/skills-manifest.json b/skills-manifest.json
index 1893e77c948..1bab6bc8b50 100644
--- a/skills-manifest.json
+++ b/skills-manifest.json
@@ -6,11 +6,11 @@
       "files": 2
     },
     "hyperframes": {
-      "hash": "7a5df5b496053c19",
+      "hash": "b222ac5809fd2633",
       "files": 77
     },
     "hyperframes-cli": {
-      "hash": "85e5d50502e7ff08",
+      "hash": "c7d7fb8c6106d8dd",
       "files": 13
     },
     "hyperframes-registry": {
@@ -18,7 +18,7 @@
       "files": 13
     },
     "media-use": {
-      "hash": "d93e5ec676b0d42b",
+      "hash": "1082bd1c4e3f033b",
       "files": 116
     },
     "remotion-to-hyperframes": {
diff --git a/skills/hyperframes-cli/references/cloudrun.md b/skills/hyperframes-cli/references/cloudrun.md
index 3ddf5396a7d..083a71b625f 100644
--- a/skills/hyperframes-cli/references/cloudrun.md
+++ b/skills/hyperframes-cli/references/cloudrun.md
@@ -22,7 +22,7 @@ npx hyperframes cloudrun destroy --project <gcp-project>
 
 `sites create` uploads a content-addressed project archive for reuse by `render-batch`; pass its `--site-id` to that command to skip the batch upload. Single `cloudrun render` currently resolves the project from its directory and does not consume `--site-id`. Both render commands require `--width` and `--height`; supported output formats are `mp4`, `mov`, `png-sequence`, and `webm`. Use `--output-resolution 4k` to supersample an authored composition without changing its layout dimensions.
 
-Common render flags are `--fps 24|30|60`, `--quality draft|standard|high`, `--codec h264|h265` for MP4, `--chunk-size`, `--max-parallel-chunks`, `--target-chunk-frames`, `--render-id`, `--output-key`, `--wait`, and `--wait-interval-ms`. Use `--json` for machine-readable output.
+Common render flags are `--fps 24|25|30|60`, `--quality draft|standard|high`, `--codec h264|h265` for MP4, `--chunk-size`, `--max-parallel-chunks`, `--target-chunk-frames`, `--render-id`, `--output-key`, `--wait`, and `--wait-interval-ms`. Use `--json` for machine-readable output.
 
 For a variable-driven single render:
 
diff --git a/skills/hyperframes-cli/references/lambda.md b/skills/hyperframes-cli/references/lambda.md
index 4d1d5cbd8be..69943bab127 100644
--- a/skills/hyperframes-cli/references/lambda.md
+++ b/skills/hyperframes-cli/references/lambda.md
@@ -84,7 +84,7 @@ Starts a Step Functions execution. Returns immediately with a `renderId` unless
 | ----------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
 | `--width` / `--height`  | Output dimensions in pixels                                                                                                                                                                                                                                                                                                                                   |
 | `--output-resolution`   | Supersampling preset (engages Chrome `deviceScaleFactor`) — `landscape` / `landscape-4k` / `portrait` / `portrait-4k` / `square` / `square-4k`, plus aliases (`1080p`, `4k`, `uhd`, `hd`, `1080p-portrait`, `4k-portrait`, `1080p-square`, `4k-square`). Use this to render an authored-at-1080p composition at 4K without re-laying-out — see footgun below. |
-| `--fps`                 | `24` / `30` / `60`                                                                                                                                                                                                                                                                                                                                            |
+| `--fps`                 | `24` / `25` / `30` / `60`                                                                                                                                                                                                                                                                                                                                     |
 | `--format`              | `mp4` / `mov` / `png-sequence` / `webm` (default `mp4`)                                                                                                                                                                                                                                                                                                       |
 | `--codec`               | `h264` / `h265` (mp4 only)                                                                                                                                                                                                                                                                                                                                    |
 | `--quality`             | `draft` / `standard` / `high`                                                                                                                                                                                                                                                                                                                                 |
diff --git a/skills/hyperframes-cli/references/lint-validate-inspect.md b/skills/hyperframes-cli/references/lint-validate-inspect.md
index 51a9b3ab002..784e3860fbb 100644
--- a/skills/hyperframes-cli/references/lint-validate-inspect.md
+++ b/skills/hyperframes-cli/references/lint-validate-inspect.md
@@ -111,7 +111,7 @@ Drop a `*.motion.json` sidecar next to the composition (matching the html basena
 npx hyperframes snapshot                       # 5 key frames as PNG
 npx hyperframes snapshot ./my-project          # specific project
 npx hyperframes snapshot --frames 10           # evenly-spaced N frames
-npx hyperframes snapshot --range 3:5 --frames 8 # 8 frames within root-timeline seconds 3–5
+npx hyperframes snapshot --range 3,5 --frames 8 # 8 frames within root-timeline seconds 3–5
 npx hyperframes snapshot --composition intro --frames 8 # 8 frames within this data-composition-id
 npx hyperframes snapshot --review              # beat sheet, seam strips, phone sheet
 npx hyperframes snapshot --review --json       # same, with paths and times as JSON
@@ -119,7 +119,7 @@ npx hyperframes snapshot --review --json       # same, with paths and times as J
 
 Still PNGs. Each run writes `.hyperframes/artifacts/snapshots/<id>/` unless `--output` is set. `snapshot` is the standalone capture; `check --snapshots` is the gate.
 
-One of `--at`, `--range START:END` (seconds), or `--composition <data-composition-id>` (not a CSS selector). `--frames` defaults to 5. Range and composition sample just before the end and add no extra end frame; `--at` adds one unless `--no-end`.
+One of `--at`, `--range START,END` (seconds), or `--composition <data-composition-id>` (not a CSS selector). `--frames` defaults to 5. Range and composition sample just before the end and add no extra end frame; `--at` adds one unless `--no-end`.
 
 `--review` picks the frames that show a film over time. Scenes are the composition's top-level timed clips and sub-composition hosts, as `timeline` lists them (captions and audio excluded); `--composition` scopes to one. It writes `review-beats.jpg` (each scene at its start + 0.3 s and at its middle, and each timed `data-beat="name"` element at its start + 0.3 s, labelled with time and id), `review-seam-NN.jpg` (frames at −0.2, −0.1, 0, +0.1, +0.2 s around every cut) and `review-phone.jpg` (the beat frames 360 px wide, to judge legibility at phone size), beside the `beat-*.png` and `seam-*.png` stills. With fewer than two scenes the beats spread evenly. `--json` prints `{ok, outputDir, frames[{path,time,role,label}], contactSheets, review{window, scenes, beats, seams[{time,from,to,times,paths,sheet}], sheets, notes}}`.
 
diff --git a/skills/hyperframes-cli/references/preview-render.md b/skills/hyperframes-cli/references/preview-render.md
index 9c25b3dbafa..128844c4aa5 100644
--- a/skills/hyperframes-cli/references/preview-render.md
+++ b/skills/hyperframes-cli/references/preview-render.md
@@ -120,7 +120,7 @@ npx hyperframes render --docker                       # byte-identical
 | `dir` (positional)                   | path                                                                                               | cwd                            | Project directory. Omit to use current working directory.                                                                                                                                                                                                                                                                                                                                                                                     |
 | `--composition`, `-c`                | path to composition file                                                                           | `index.html`                   | Render a specific composition file (e.g. `compositions/intro.html`) instead of the project's `index.html`.                                                                                                                                                                                                                                                                                                                                    |
 | `--output`, `-o`                     | path                                                                                               | `renders/<project>_<ts>.<ext>` | Output path. Default is timestamped (`<project-name>_YYYY-MM-DD_HH-MM-SS.<ext>`).                                                                                                                                                                                                                                                                                                                                                             |
-| `--fps`                              | 24, 30, 60                                                                                         | 30                             | 60fps doubles render time                                                                                                                                                                                                                                                                                                                                                                                                                     |
+| `--fps`                              | 1–240, or a rational such as `30000/1001`                                                          | root `data-fps`, else 30       | 60fps doubles render time. `lambda` and `cloudrun` accept only 24, 25, 30, 60                                                                                                                                                                                                                                                                                                                                                                 |
 | `--quality`                          | draft, looks, delivery, standard, high                                                             | looks                          | `looks` is CRF 16 on the standard preset. `delivery` is `high`. draft for iterating                                                                                                                                                                                                                                                                                                                                                           |
 | `--format`                           | mp4, webm, mov, gif, png-sequence, hls                                                             | mp4                            | WebM/MOV render with transparency; gif for inline autoplay in GitHub PRs/READMEs/docs (two-pass palette encode, fps capped at 30 — prefer `--fps 15` — no audio, 1-bit transparency only, HDR falls back to SDR); png-sequence writes RGBA frames to a directory (AE/Nuke/Fusion ingest); hls writes an HLS VOD directory (master.m3u8 + video/audio playlists + MPEG-TS segments), SDR only, rejects `--gpu`, unavailable on lambda/cloudrun |
 | `--gif-loop`                         | 0-65535                                                                                            | 0                              | GIF loop count; `0` loops forever. Only with `--format gif`.                                                                                                                                                                                                                                                                                                                                                                                  |
diff --git a/skills/hyperframes/references/captions.md b/skills/hyperframes/references/captions.md
index ac8e8e5eac4..db13fc7e3ed 100644
--- a/skills/hyperframes/references/captions.md
+++ b/skills/hyperframes/references/captions.md
@@ -3,7 +3,7 @@
 ## Words and timings
 
 - `npx hyperframes transcribe <file> --json` and a HeyGen voiceover (`/media-use`) give every word's start and end. Work from those as `[{ "id": "w0", "text": "Hello", "start": 0.0, "end": 0.5 }, …]`.
-- Local transcription defaults to `small.en`, which **translates** non-English speech into English. Pass the model: `npx hyperframes transcribe <file> --model small.en` for English, `--model small --language <code>` otherwise, `--model medium.en` for speech over music. It writes `assets/transcript.json` (its JSON names the path) and overwrites it on the next run; `--to srt` or `--to vtt` writes a sidecar.
+- `npx hyperframes transcribe` defaults to the `small.en` whisper model, which **translates** non-English speech into English. Pass the model: `npx hyperframes transcribe <file> --model small.en` for English, `--model small --language <code>` otherwise, `--model medium.en` for speech over music. It writes `assets/transcript.json` (its JSON names the path) and overwrites it on the next run; `--to srt` or `--to vtt` writes a sidecar.
 - For Chinese, Japanese and other unspaced scripts, local transcription returns a whole sentence as one "word", sometimes in the other script (Traditional for Simplified speech). Split each segment's text into short phrases and spread its time across them by character count; check the characters against what is said.
 - Whisper invents "Thank you." over silence and repeats lines over a silent tail: drop words that start inside a final silence (`silencedetect=noise=-35dB:d=0.6`; real speech sits around −16 to −26 dB).
 - Never put two words in one timed entry, and match transcript words to caption words by position, not text, or a repeated word takes the wrong time.
diff --git a/skills/media-use/audio/references/transcribe.md b/skills/media-use/audio/references/transcribe.md
index 06df2ce445b..59d9c488617 100644
--- a/skills/media-use/audio/references/transcribe.md
+++ b/skills/media-use/audio/references/transcribe.md
@@ -1,6 +1,6 @@
 # Transcription
 
-Create normalized word-level timestamps. **Always specify `--model` explicitly** — the CLI default is `small.en`, which silently translates non-English audio into English.
+Create normalized word-level timestamps. **Always specify `--model` explicitly** — `npx hyperframes transcribe` uses `small.en` as its default whisper model, which silently translates non-English audio into English.
 
 ```bash
 npx hyperframes transcribe audio.mp3  --model small.en             # known English
@@ -19,7 +19,7 @@ npx hyperframes transcribe openai-response.json
 2. **Known non-English** → `--model small --language <iso-code>` (no `.en` suffix)
 3. **Unknown language** → `--model small` (whisper auto-detects)
 
-**CLI default is `small.en`** — do not rely on it; always pass `--model` to make the choice explicit. `--language` also filters out non-target-language segments from mixed-language audio.
+**The CLI's default whisper model is `small.en`** — do not rely on it; always pass `--model` to make the choice explicit. `--language` also filters out non-target-language segments from mixed-language audio.
 
 ## Model Sizes
 
diff --git a/skills/media-use/references/audio.md b/skills/media-use/references/audio.md
index 4ffce7a3d0e..350185512b9 100644
--- a/skills/media-use/references/audio.md
+++ b/skills/media-use/references/audio.md
@@ -17,4 +17,4 @@ node <SKILL_DIR>/audio/scripts/audio.mjs --request ./audio_request.json --out ./
 
 Single-shot helpers: `audio/scripts/heygen-tts.mjs` (one voice file). Transcription / background removal / captions use the `hyperframes` CLI (`transcribe`, `remove-background`), see the per-topic guides in `audio/references/` (`tts.md`, `bgm.md`, `sfx.md`, `transcribe.md`, `remove-background.md`, `captions/`).
 
-Transcription defaults to Parakeet (better than whisper.cpp: 6.05% vs 7.44% WER, 5-10x faster) via `scripts/transcribe.mjs`, with whisper.cpp auto-fallback (see `references/operations.md`).
+Transcription through `scripts/transcribe.mjs` defaults to Parakeet (better than whisper.cpp: 6.05% vs 7.44% WER, 5-10x faster), with whisper.cpp auto-fallback (see `references/operations.md`). `npx hyperframes transcribe` picks its own engine: Parakeet when installed, otherwise whisper with the `small.en` model by default.