embedded-captions

Add captions or subtitles to an existing single-subject talking-head video without editing the footage. Use for plain verbatim captions, cinematic captions embedded behind the subject, VFX captions, “炸/特效/酷炫字幕,” or a named identity from the 35-style catalog. Route by visual identity, not by backend engine. The quiet `anchor` rail is the default; embed every word only when the user explicitly wants a fully cinematic treatment. The workflow runs locally end to end, including transcription and subject matting; split multi-shot footage before applying it.

Quality

86%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

73%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.

The content is a well-structured, highly actionable overview of a complex multi-mode skill with strong validation and feedback loops. Its main weakness is verbosity from restating core rules across multiple sections.

Suggestions

Consolidate the rail/embed/drop model into the Caption model section and have the intro, Load-bearing rules, and Non-negotiables link back to it rather than re-explaining.

Trim the dense parenthetical asides in the Operational flow and Non-negotiables into short rules with detail moved to the relevant reference file.

Move the full DNA registry table and per-identity detail fully behind dna/README.md, keeping only the pick heuristic and a one-line-per-DNA summary in SKILL.md.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient craft detail, but the rail/embed/drop model is restated across the intro, Caption model, Load-bearing rules, and Non-negotiables, and several long parenthetical asides could be tightened.	3 / 5
Actionability	Provides concrete, copy-paste-ready commands (prepare.sh, preview-frames.cjs, render-and-composite.sh, ffprobe/ffmpeg probes), but full authoring schemas for plan.json/theme.json are deferred to external files, leaving minor gaps.	4 / 5
Workflow Clarity	A clearly sequenced 5-step pipeline with a 'RUN FIRST' decision gate, explicit validation gates (timing/occlusion/overflow/hand-off), and a Visual QA checklist plus a preview→fix→re-preview→render feedback loop for error recovery.	5 / 5
Progressive Disclosure	A Shared-knowledge table gives clear one-level-deep navigation to real reference files (rail.md, composition-craft.md, dna/README.md, etc.), though the main SKILL.md itself is long with some inlined detail that leans on reinforcement.	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.

The description is specific, trigger-rich, and clearly answers both what and when in third-person imperative voice. It is a strong, concise pitch with minimal overlap risk.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — plain verbatim captions, cinematic captions embedded behind the subject, VFX captions, and named identities from a 35-style catalog — plus transcription and matting, giving comprehensive coverage of the skill's capabilities.	5 / 5
Completeness	Explicitly states both what it does ('Add captions or subtitles to an existing single-subject talking-head video without editing the footage') and when to use it ('Use for plain verbatim captions, cinematic captions…, VFX captions…').	5 / 5
Trigger Term Quality	Includes natural user-voiced terms and synonyms — 'captions or subtitles', 'cinematic captions', 'VFX captions', and the Chinese buzzwords '炸/特效/酷炫字幕' — covering the phrases users actually say.	5 / 5
Distinctiveness Conflict Risk	Tightly scoped to single-subject talking-head video captioning with distinct visual-identity routing, making conflict with other skills minimal.	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
relative_links	Relative link issues: 12 missing	Warning

	Total	15 / 16 Passed

Repository: heygen-com/hyperframes
Path: skills/embedded-captions/SKILL.md
Commit: a562946

Reviewed: about 11 hours 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.