video-transcripts

Generate structured video transcripts from local files or video URLs using Gemini Files API. Use when a GitHub or Linear tracker item, comment, or attachment includes a screen recording, .mov, .mp4, or tracker-hosted video and you need a <video-transcripts> block instead of hand-written notes.

Quality

88%

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

87%

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

A high-quality, actionable skill body with copy-paste-ready commands, a precise output contract, and clean one-level-deep bundle usage. The main gap is that validation/feedback-loop checkpoints live inside the helper script rather than being made explicit in the workflow.

Suggestions

Surface an explicit validation/feedback loop in the Workflow (e.g. "If the transcript is malformed or low-signal, re-run with --model gemini-3-flash-preview or --debug-dir and retry"), so checkpoints are visible to Claude rather than hidden in the script.

Consolidate the lightly overlapping rules (e.g. step 10 against the Output Contract's "Do not invent hidden state") to trim a few near-duplicate tokens.

Dimension	Reasoning	Score
Conciseness	The body is lean and concrete — executable bash commands, an exact XML contract, and a model strategy — with no padding explaining concepts Claude already knows; nearly every token earns its place despite a few near-redundant workflow rules.	3 / 3
Actionability	Three copy-paste-ready bash invocations with real URL/path/title arguments, env-var and flag overrides, the exact output XML shape, and the exact cache-comment format make the guidance fully executable and specific.	3 / 3
Workflow Clarity	The 17-step Workflow is clearly sequenced with cache-matching logic, but validation checkpoints are implicit (delegated to the helper script's quality gate) rather than surfaced as an explicit in-body validate→fix→retry loop, which caps it per the batch/risky-operations guideline.	2 / 3
Progressive Disclosure	Content is well-organized into clear sections and references a single real one-level-deep bundle file (scripts/generate_video_transcript.sh, confirmed to exist) by concrete path, with no nested references.	3 / 3
	Total	11 / 12 Passed

Description

90%

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, third-person description that clearly pairs a concrete capability with explicit, natural-language use-when triggers anchored to specific tracker attachment types. It would benefit only from naming a couple more concrete actions to reach full specificity.

Suggestions

Add one or two more concrete actions (e.g. "extract a timeline of observed actions" or "normalize tracker-hosted URLs") to lift specificity from a single core action toward multiple concrete capabilities.

Consider mentioning the output shape briefly (e.g. "emits an XML <video-transcripts> block") in the description so users searching for that artifact type also match.

Dimension	Reasoning	Score
Specificity	Names one concrete action ("Generate structured video transcripts") with specific input sources (local files / video URLs via Gemini Files API), but does not list multiple distinct concrete actions like the score-3 anchor, so it sits at "names domain and some actions, not comprehensive."	2 / 3
Completeness	Explicitly states what it does (generate structured video transcripts) and when to use it via an explicit "Use when" clause tied to specific attachment types, answering both what AND when.	3 / 3
Trigger Term Quality	Covers natural terms a user would say — "screen recording", ".mov", ".mp4", "video", plus GitHub/Linear tracker contexts — giving good coverage of natural variations rather than a single keyword.	3 / 3
Distinctiveness Conflict Risk	The niche is narrow (Gemini Files API video transcripts for GitHub/Linear tracker attachments) with triggers tied to specific hosts and file types, making it unlikely to fire for unrelated skills.	3 / 3
	Total	11 / 12 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: 1 missing, 2 suspicious	Warning

	Total	15 / 16 Passed

Repository: udecode/plate
Path: .agents/skills/video-transcripts/SKILL.md
Commit: 45116fd

Reviewed: about 13 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.