* fix(engine): short clips from Apple-encoded videos keep their last frame
* refactor(engine): name the end-of-source hold and frame bound instead of commenting them
* fix(engine): frame ranges past a source's end stay empty with the held final frame
* test(engine): move the final-frame test into its own file
* test(engine): nightly motion blur run over generated compositions
A seeded generator of text-free compositions combines 19 features (nested
canvas, overflow, clip path, scroll container, callbacks, a last-child rule)
and checks the two pixel invariants on auto: outside the trail at two holds
and at a moving time, inside the trail at a moving time. The runtime lane
runs seeds 1 to 3; motion-blur-nightly.yml runs 200 a day on the Chrome 153
pin and on stable, and opens one issue per failing seed.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* test(engine): the nightly records every failure, ignores noisy references, caps its issues
A crash, a timeout or a launch failure is recorded before the job goes red; a
failing check whose blur-off reference disagrees with itself is reported as
unstable instead of failing; the report job moves into a tested script that
fetches open titles once, caps first, and opens one systemic issue above 5 seeds.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* test(engine): the nightly report resists injected failure text and keeps a crashed leg's record
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* fix(engine): nightly report escapes backslashes, test strips tags to a fixpoint
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* fix(cli): templates pull unpacks under .hyperframes so a watched project can take the template
* fix(cli): templates pull's staging comment fits on one line
createSvgContactSheet built each path with path.join and then took the
label with split("/"), so on Windows the label was the whole path and
the 60-character cell limit cut the file name off. Use basename, as the
scroll and snapshot sheets in the same file already do.
* fix(cli): staging folders beside a project path no longer fail past 260 characters on Windows
mkdtemp does not get the long-path prefix that mkdir does, so every staging folder
made beside a project or output path failed once its full path passed 260 characters.
They now come from one helper, createStagingDir in @hyperframes/core/atomic-file, and a
scan test fails on a new mkdtemp beside a path.
* fix(cli): give mkdtemp the long-path form instead of replacing it
CodeQL treats mkdtemp as the safe way to make a folder under the system temp dir and
flagged every file written into a folder from the mkdir-based helper. createStagingDir
now calls mkdtemp with the long-path form of the path on Windows, so both hold.
* fix(cli): leave non-Windows paths alone when making a staging folder
The long-path form is now chosen by the shape of the path, not process.platform,
so tests that pretend to be Windows keep working and the comment share does not rise.
* fix(core): a staging folder keeps a Windows path that already names its namespace
* fix(core): a staging folder keeps a Windows device path as the caller gave it
* fix(core): the staging folder helper's comment fits the line limit
* fix(cli): no macOS developer-tools prompt from git, xcrun or cc checks on a Mac without the tools
* fix(cli): treat Xcode.app as installed tools and only take executables from PATH
* test(cli): give skills tests a developer folder so darwin cases run on any host
* fix(cli): read macOS tool paths the same way on every host
Follow-ups from review on #7443 and #7444:
- The quote hint returned the first run of words that existed, so a
"Final Cut/" folder hid "Final Cut export.mp4". It now names the longest.
- "--compare=my clip.mp4" got no hint; the value after = now starts a run.
- The telemetry section lists the skill a reverse or templates command ran
for, and the HeyGen session id a recorded reverse run carries.
* fix(core): detect template entries structurally
isTemplateEntry only matched a raw file that starts with <template>, so the
full-document sub-composition shape (<!doctype html><html><body><template>
<div data-composition-id>...</template></body></html>) was not recognised and
resolveEntryHtmlPath compiled the raw file, whose root sits in an inert
template: no timeline, no duration, "Composition has zero duration".
A template entry is now a document whose <template> holds the composition
root with no live composition root outside it. Detection and extraction share
one lookup (locateSceneRoot), the same one readSceneRoot uses.
— Jerrai
* fix(core): drop comments from the template lookup
The comment ratchet allows no net rise in a file's comment share; the PR
description carries the explanation.
— Jerrai
* fix(core): find the scene template among several templates
locateSceneRoot read only the first <template>, so a scene template that
follows another template was treated as not a template entry. Scan every
template and take the first whose content holds a composition root.
* fix(core): drop the added comment (comment ratchet)
* chore: raise the inspector unpacked size budget by 0.5%
The inspector bundles core; the multi-template scene lookup added a few
hundred bytes and put it 10 bytes over the 1% tolerance.
* fix(core): refuse a file with several scene templates, naming them
* chore: trim the inspector size budget bump to what is needed
* fix(core): render the first scene template and lint the ambiguity instead of failing
A file with several scene templates and no live root rendered its first one
before; locateSceneRoot threw and stopped such projects rendering. It returns
the first again, and the multiple_scene_templates lint warning names each.
Co-authored-by: Jerrai <jerrai@heygen.com>
* chore(core): drop the locateSceneRoot comment line the comment ratchet counts
Co-authored-by: Jerrai <jerrai@heygen.com>
* fix(core): one shared scene-template pick for the compiler, the inliner and the loader
A scene template after a card template was detected by locateSceneRoot but
mounted from the first template of any kind, so it rendered blank.
findSceneTemplate picks the first template holding a composition root and
locateSceneRoot, resolveSubCompositionContent and the loader all use it.
Co-authored-by: Jerrai <jerrai@heygen.com>
* feat(cli): reverse and templates report their own runs; commands say which skills are installed
Gaps in what skill telemetry can answer for /reverse-any-video and
/style-sampling:
- The skill's beacon fires only when the agent runs it, and agents often
skip it. `reverse start` and `templates pull` now send `skill_invoked`
with `source: command`; beacons carry `source: beacon`. A person counts
once.
- style-sampling renders were never tagged. A pull marks the project the
episode is built in (only a folder that is already a project).
- Installs of an on-demand skill were invisible. `cli_command` carries
`skills_installed` and `skills_outdated`, the names from the cached
background skills check, so "never triggered" and "not installed" can
be told apart.
- `auth status` signed out exits 1 with a bare "Command failed", which
reads like a login failure. It now carries `NotSignedIn`, and an auth
error its own cause.
And a step that gets lost: a reverse started signed out often goes on
without ever trying to sign in. `reverse start` signed out now prints
`login_command` and a `next` line in the JSON the agent reads, not only
in the skill's reference.
* refactor(cli): keep comment share at the base in the reverse telemetry files
* fix(cli): command runs get their own event, counted once per piece of work
Review on the first cut: `skill_invoked` from the commands doubled the
#7176 beacon's single observation, and it fired before any work.
- `reverse start` and `templates pull` send `skill_command_ran`, not
`skill_invoked`; the beacon is back to its #7176 shape.
- `reverse start` sends it after the probe, with `signed_in` and the
reverse `session_id` that keys a retry; a video it cannot probe sends
nothing.
- `templates pull` sends it only when it downloads; a re-pull or a
refused pull sends nothing.
- A skills check that finds no install forgets the installed names.
- `auth status` failures carry only a code, never the auth error's text.
* fix(cli): measure answers the four mistakes agents make most
The usage errors agents hit most on `measure`: no op named, a path it
cannot read, and a flag it does not take. Some agents never get past
them to a successful measure.
- `measure <video>` with no op (alone or with --json) runs `probe` and
says so on stderr. With other flags the op is a guess, so the
unknown-command hint asks for one; it now names a path that does not
exist yet too.
- When ffprobe cannot read a file, the error carries ffprobe's reason and
a matching fix: a cut-short download, a page saved as .mp4, a
permission.
- An unquoted path with spaces says to quote it, both where the extra
words are rejected and where the cut path is missing.
- Window flags take --from/--to, and folder flags take -o.
* refactor(cli): keep comment share at the base in the measure files
* test(cli): expect a spaced path as the shell split it, on Windows too
* fix(cli): a cut-short MP4 says download it again, read from its header and ffprobe's specific line
* fix(cli): an unquoted spaced path reaches probe or its quote hint wherever it falls
* feat(skills): reverse-any-video and style-sampling install with the core set
A skill only triggers when it is installed. On demand these two rarely
were, so a request to copy a video or make an episode in a proven style
seldom reached them, and the entry skill named only one of them, in its
closing line.
- isCoreSkill, FALLBACK_CORE_SKILLS and the marketplace `core-skills`
entry include both, so `init` and `skills update` install them. The
`hyperframes` plugin already bundled every skill.
- `/hyperframes` routes to both from the top of its table: a video to
copy or remake, and an episode in a proven style. Its budget takes
+64 B for those two rows.
- `/reverse-any-video` triggers on "remake" and on 「复刻这个视频」
「照着做一个」, inside the 320 B description cap.
- Every catalog surface lists both in the core group: CLAUDE.md,
AGENTS.md, README, the skills guide, skill-lifecycle and the two init
templates (still byte-identical).
* refactor(cli): keep comment share at the base in skillsManifest
* fix(core): a video held on its tail is seeked once, not every frame
* fix(core): seek a held tail again when the element was moved off it
* fix(core): the held-tail skip applies only while playing
* style(core): format the held-tail change
* refactor(core): drop a comment that repeats its neighbour
* feat(producer): distributed and range renders honour page motion blur
A film with data-motion-blur="on" blurs the same way however it is rendered. plan() no longer
forces the synthetic job to pageMotionBlur "off", so the plan locks screenshot capture for a
blurred film as a whole render does, and renderChunk() no longer defaults the page's off switch
to "distributed-render". An explicit pageMotionBlur "off" in the config still stands the page down.
Hosted renders pick this up after the next OSS release and a producer pin bump; a render of a
blur-on film that goes distributed changes output then.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* test(producer): the default tier blurs the same in chunks; docs agree distributed renders blur
Two files compare plan -> renderChunk -> assemble with the tier attribute removed, so `auto`
(the filter tier) decides, against a whole render: held-card and callbacks (the muted-callback
path). The skill's motion reference no longer says distributed renders are off. The rewritten
R19 plan test gets 60 s for its three plan() calls.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
CLAUDE_CODE_ENTRYPOINT is set by current Claude Code versions too, and it
is the only Claude marker MCP servers inherit, so it is not an
"older versions" fallback. The Agent SDK comment still said Claude Code's
shell tool sets CLAUDECODE; the rule keys on CLAUDE_CODE_CHILD_SESSION.
Follow-up to review nits on #7425. Comments only.
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* feat(cli): add agent markers other detectors use, verified upstream
Compared std-env, @vercel/detect-agent, am-i-vibing and agent-cli-detector
against our rules and kept only markers the agent itself sets on the
shells it spawns: Auggie AUGMENT_AGENT, Amp AMP_CURRENT_THREAD_ID, Cursor
CURSOR_CONVERSATION_ID and CURSOR_SANDBOX, Cline CLI CLINE_WRAPPER_PATH,
Muse MUSE_RELEASE_INFO, and Warp Oz OZ_RUN_ID (after the harnesses it
wraps).
Claude Code now keys on CLAUDE_CODE_CHILD_SESSION or
CLAUDE_CODE_ENTRYPOINT. Its IDE extensions also set CLAUDECODE in the
integrated terminal a person types into, so CLAUDECODE alone counted that
person as an agent: no prompts, agent-mode login.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* docs(cli): trim comments and fix citations the comment gates flag
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* chore: raise the cli packed size budget to its measured size
main already sat at the edge of the 1% tolerance; the new agent rules
put the packed CLI 842 bytes past it (8062664 vs 7982002 + 1%).
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
TERM_PROGRAM=cursor|windsurf and Replit's REPL_ID only say where a shell
runs. They now sit in their own list, checked after the AI_AGENT / AGENT
name check, so Amp in a Cursor terminal or Devin in Replit report as
themselves, and an unknown AI_AGENT there keeps its discovery hints.
agentOnly had no users left, so it is gone.
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* feat(cli): detect more agent runtimes in telemetry
Agent runtime rules:
- copilot_agent: key on COPILOT_AGENT=1 or AI_AGENT=github_copilot_*
(VS Code agent terminals) instead of the unverified Actions markers.
- copilot_cli: COPILOT_CLI=1 or COPILOT_AGENT_SESSION_ID, ahead of
copilot_agent.
- kilo (KILO=1, KILOCODE_VERSION), ahead of opencode.
- opencode (OPENCODE=1, OPENCODE_SESSION_ID).
- qwen_code (QWEN_CODE=1), junie (JUNIE_SHIM_PATH),
antigravity (ANTIGRAVITY_AGENT), qoder (QODER_AGENT),
kiro (AGENT_DISPLAY_OUT, AGENT_CONTEXT_OUT), openhands (AI_AGENT=openhands).
- AI_AGENT fallback after every vendor rule: known names map to their
runtime, any other slug-safe name becomes other_ai_agent. Bare AGENT
is not used.
Discovery hints: other_ai_agent keeps its hints, agent_hint prefers
AI_AGENT over AGENT, and AGENT_TOOLSDIRECTORY,
GSM_SKIP_SSH_AGENT_WORKAROUND and EDGE_METRICS_SESSION_ID are no
longer hint keys.
CI: name AWS CodeBuild (CODEBUILD_BUILD_ID) as codebuild.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* feat(cli): detect agents seen in discovery hints
Add rules for closed-source agents whose markers show up in the CLI's
own agent_env_hints on non-interactive runs: WorkBuddy, CodeBuddy, Trae,
Pippit, Doubao, Aily, Grok, Kimi, ZCode, SWE-agent and apps built on the
Claude Agent SDK. Qoder also matches its CLI and agent SDK keys.
The AI_AGENT name map now covers the names agents report (Devin, Amp,
v0, Rovo Dev, Base44, Trae, Antigravity, OpenHands, Copilot, and
hermes-agent), and a known name in the bare AGENT variable counts too.
Unknown AGENT values such as AGENT=1 still do not.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* feat(cli): detect the DeepSeek harness
Every dsh model shell call runs with DSH_SHELL=1 and the agent's
DSH_SESSION_ID. HyperFrames Desktop also runs dsh; its runs carry
HYPERFRAMES_CLIENT, so the client field separates the two.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* docs(cli): cite the Agent SDK source for claude_agent_sdk
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* fix(cli): let the closest agent win over inherited markers
Copilot's COPILOT_AGENT / AI_AGENT and COPILOT_CLI pass down to the
agents it starts, and TERM_PROGRAM / REPL_ID only say where a shell runs.
Those rules now come after every other agent rule, so opencode or Qwen
started by Copilot, or Kilo in a Cursor terminal, report as themselves.
dsh keys on DSH_SHELL=1 only: it sets DSH_SESSION_ID on the terminal a
person types into too. Qoder drops its app-wide SDK and setting keys,
CodeBuddy drops CODEBUDDY_AGENT_RUNTIME (a setting it reads), and Kiro's
markers must be non-empty.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* feat(cli): add 1440p resolution presets to --resolution
Adds landscape-1440p (2560x1440), portrait-1440p (1440x2560) and
square-1440p (1440x1440) next to the existing 4K, 720p and 480p tiers,
with the aliases 1440p, 2k and qhd plus 1440p-portrait and 1440p-square.
1440p joins the aspect-agnostic aliases, like 720p and 480p.
Every validator derives from CANVAS_DIMENSIONS, so no logic changes: a 4K
composition renders down to 1440p through the existing smaller-preset
path, and a 1280x720 composition supersamples to 1440p at an exact 2x.
A 1920x1080 composition is still refused by the non-integer-scale check
(4/3); fractional scale factors are not part of this change.
Refs #6309
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* fix(cli): clarify 1440p aliases and repair skill validation
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
Co-authored-by: James <james.russo@heygen.com>
* fix(cli): cloud render finds media whose file name is stored in another Unicode form than the page's
* fix(cli): keep the archive-key change from raising publishProject's comment share
* fix(cli): a page that names a decomposed file as the disk lists it still matches its archive entry
* perf(studio-server): a large project's first history copy takes a fraction of the time
* perf(studio-server): read a small file through one handle, with fewer comments
* fix(studio-server): small files keep their exact permissions in history under a strict umask
The declaration rule validates data-composition-variables on every
declaring element since 571c99ce9 (#4454), but the template-fragment
reproduction case from #4383 had no regression coverage. Add it.
Refs #4383
Co-authored-by: factory-droid[bot] <138933559+factory-droid[bot]@users.noreply.github.com>
* fix(core): a muted video's picture no longer waits on its own audio track at a cut
A muted video takes its clock from its audio track in Safari; a late start of that track stalls the picture after the cut. The runtime now switches the track off.
* test(core): a video muted by script and unmuted later keeps its audio track
* fix(core): type the audio track loop so it passes the strict index check
* fix(producer): every render draws the same random values, at any worker count
* fix(producer): run the random-stagger fixture in CI and say what seeding covers
* feat(cli): check returns an unchanged project's last pass in about a second
* fix(cli): the check cache hashes files without a stat race and its tests run before the signal test
* test(cli): the warm check test compares two real runs, not a run and its cached pass
* feat(cli): check --no-craft skips the craft pass for a render preflight
* fix(cli): the check cache keys on timeout, env and large files and never replays a moved tree
* style(cli): format the check cache
* fix(cli): the large-file cache test waits for the clock tick it depends on
* fix(cli): the check cache keys on the browser and on scripts under .hyperframes
* fix(cli): the check cache ignores the video copies the check itself touches
* fix(core): a set written before its tween no longer erases the tween's end
* fix(core): shorten two comment lines to the limit
* fix(core): the stamp takes any GSAP the runtime types, so the runtime typecheck passes