captions-overlay

Overlay doctrine for the embedded-captions workflow — the caption MODEL (drop / rail / embed) and the rule that captions are an OVERLAY composited on top of the film, never a reserved bottom band you shift content up to avoid. Load when adding captions/subtitles to a talking-head or launch video, when deciding whether a phrase should be dropped, ride the verbatim rail, or be promoted to a scarce embedded climax, when laying out a composition that will carry captions (do NOT reserve a keep-out band), or when centering a composition on the true frame center under captions. Quotes the rail+embed model from embedded-captions and constraint

Quality

79%

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

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/captions-overlay/SKILL.md

Quality

Content

67%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 well-organized doctrine body with unusually concrete numeric rules (center coordinates, canvas percentages, embed scarcity/spacing) and a correct overlay-not-keep-out framing throughout. Its main cost is redundancy — the closing synthesis section and the rail-first bullets restate content already given — and the absence of a worked example showing the doctrine applied to a real phrase.

Suggestions

Cut or compress the "Why these two rules are one doctrine" section (and the overlapping rail-first bullets) since they restate the intro and table, tightening conciseness toward a lean body.

Add one short worked example — a transcript phrase or two classified as drop / rail / embed with the resulting composition choice — to make the actionability fully concrete without relying on the upstream skill.

Replace the inline quoted upstream non-negotiables with a single clear pointer to the `embedded-captions` skill (or a local reference file) so quoted material is not duplicated inside the body.

Dimension	Reasoning	Score
Conciseness	Mostly efficient, domain-specific rule content with concrete numbers (y = H/2, landscape 540 / portrait 960, bottom ~5-8%, ~80px center span), but the closing "Why these two rules are one doctrine" section restates the intro nearly point-for-point and the rail-first bullets repeat the table, so it could be tightened.	3 / 5
Actionability	Concrete, executable-for-a-doctrine guidance: exact centering values ("y = H / 2 (landscape 540, portrait 960)"), embed scarcity rules ("≤1 per beat, never two adjacent/co-visible, spaced ≥ a beat apart"), and the ~80px courtesy rule; implementation detail is appropriately delegated to the upstream embedded-captions skill.	4 / 5
Workflow Clarity	Not a multi-step pipeline — each doctrine rule is unambiguous and it names a verification path ("Finalize snapshot QA judges caption-over-content legibility visually"). Falls short of 5 because the per-phrase drop/rail/embed decision procedure is left to judgment and the upstream skill rather than given as a worked sequence.	4 / 5
Progressive Disclosure	No bundle files exist (no references/, scripts/, or assets/), and the ~76-line body is organized into clear sections with a well-signaled one-level pointer to the upstream `embedded-captions` skill. Slightly over the under-50-line simple-skill bar and the quoted upstream non-negotiables are inlined rather than referenced.	4 / 5
	Total	15 / 20 Passed

Description

83%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 description: it states the doctrine's two ideas concretely, gives explicit multi-clause load triggers, and clearly delimits its relationship to the upstream embedded-captions skill. Its weaknesses are insider jargon that a user wouldn't naturally say and a second-person clause in the framing.

Suggestions

Add user-natural synonyms such as "closed captions" or "CC" to the trigger clauses to broaden trigger_term_quality.

Reword "a reserved bottom band you shift content up to avoid" into third person (e.g., "a reserved bottom band content is shifted up to avoid") to keep the description in third-person voice throughout.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions ("deciding whether a phrase should be dropped, ride the verbatim rail, or be promoted", "centering a composition on the true frame center under captions") with concrete mechanics, but leans on internal jargon ("constraint #13", "scarce embedded climax") and dips into second person ("a reserved bottom band you shift content up to avoid"), which the guidelines penalize.	4 / 5
Completeness	Explicitly answers both: the "what" (the drop/rail/embed caption model plus the overlay-not-reserved-band law) and the "when" ("Load when adding captions/subtitles to a talking-head or launch video, when deciding whether a phrase should be dropped…, when laying out a composition…, or when centering a composition…").	5 / 5
Trigger Term Quality	Good natural keyword coverage — "adding captions/subtitles to a talking-head or launch video", "captions", "subtitles", "drop / rail / embed" — but misses common variations users would say such as "closed captions", "CC", or "text on video".	4 / 5
Distinctiveness Conflict Risk	Distinct niche with explicit boundary declaration ("Applies ON TOP of embedded-captions"), but it shares trigger territory with the upstream embedded-captions skill itself, so it may fire where the upstream skill alone is wanted.	4 / 5
	Total	17 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning

	Total	14 / 16 Passed

Repository: heygen-com/hyperframes
Path: .agents/skills/captions-overlay/SKILL.md
Commit: a4a7cbe

Reviewed: 3 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.