talking-head-recut

Package an existing talking-head / interview / podcast video with timed, designed GRAPHIC OVERLAY cards — kinetic titles, lower-thirds, data callouts, quotes, side panels, picture-in-picture — synced to the transcript, on a 16:9 / 9:16 / 4:5 canvas of your choice; the clip plays untouched underneath. Trigger on "graphic overlays", "on-screen graphics", "package / dress up my video". Not plain subtitles (/embedded-captions). Unclear → /hyperframes.

Quality

87%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

75%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A highly actionable, well-sequenced skill with concrete commands, templates, and a clean reference library structure. Its main weakness is length: repeated layout/zone tables and re-stated routing blockquotes could be consolidated to respect the token budget.

Suggestions

Consolidate the zone/bounds information into one table (Step 6) and reference it from Step 7 and Step 9 instead of re-listing the GSAP target bounds three times.

Trim the opening blockquotes: the /hyperframes front-door and embedded-captions-sibling points are restated multiple times in the first ~25 lines; keep one concise statement.

Make the render validation an explicit feedback loop in Step 10 (snapshot → if wrong, fix card HTML/timeline → re-snapshot → only then full render) to lift workflow clarity.

Dimension	Reasoning	Score
Conciseness	The body is dense and mostly operational (commands, tables, templates), but at ~1188 lines it repeats information — the zone/bounds table appears in Step 6, again as 'Other zone values' in Step 7, and as GSAP targets in Step 9, and several opening blockquotes re-explain the /hyperframes routing — so it could be tightened.	3 / 5
Actionability	Fully executable guidance throughout: exact ffprobe/ffmpeg commands with flags, a complete copy-paste composition HTML template, a GSAP statement cheat sheet with per-anim templates, and a closed data-anim table with params — copy-paste ready for the common cases.	5 / 5
Workflow Clarity	A clear 11-step sequence with checkpoints (hyperframes doctor, clamp-to-duration warning, snapshot --at sanity check before full render, QA rules), but the validate→fix→retry loop is implied rather than spelled out as an explicit feedback loop.	4 / 5
Progressive Disclosure	A well-organized one-level-deep reference bundle (DESIGN_INDEX.md plus styles/layouts/frames) is clearly signaled and navigable, and all referenced paths resolve to real files; however the main body inlines a large composition template and repeated tables that keep it long.	4 / 5
	Total	16 / 20 Passed

Description

100%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, concrete description that explicitly covers what the skill does, when to trigger it, and how it differs from sibling skills. It avoids vague fluff and uses natural trigger terms a user would actually say.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete overlay actions — 'kinetic titles, lower-thirds, data callouts, quotes, side panels, picture-in-picture' — plus canvas options '16:9 / 9:16 / 4:5' and the invariant 'clip plays untouched underneath', giving comprehensive coverage of what the skill does.	5 / 5
Completeness	Clearly states what ('Package an existing talking-head ... video with timed, designed GRAPHIC OVERLAY cards ... synced to the transcript') and when ('Trigger on "graphic overlays", "on-screen graphics", "package / dress up my video"') with concrete trigger phrases.	5 / 5
Trigger Term Quality	Explicit natural triggers users would say — 'graphic overlays', 'on-screen graphics', 'package / dress up my video' — with synonyms and domain terms (talking-head / interview / podcast); file extensions don't apply to this concept-level domain.	5 / 5
Distinctiveness Conflict Risk	Clear niche (designed graphic overlays on a playing talking-head clip) with explicit disambiguation — 'Not plain subtitles (/embedded-captions)' and 'Unclear → /hyperframes' — minimizing wrong-skill triggering.	5 / 5
	Total	20 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (1219 lines); consider splitting into references/ and linking	Warning

	Total	15 / 16 Passed

Repository: heygen-com/hyperframes
Path: skills/talking-head-recut/SKILL.md
Commit: 0f8eb89

Reviewed: 6 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.