multi-video-synthesis

Performs high-density targeted extraction across 10+ videos to map semantic landscapes, consensus, and controversies.

Quality

61%

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

Fix and improve this skill with Tessl

tessl review fix ./python/agents/youtube-analyst/youtube_analyst/skills/multi-video-synthesis/SKILL.md

Quality

Content

68%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 body is an efficiently written, actionable orchestration playbook with concrete tool calls and a clear 6-step sequence. Its main weakness is the absence of validation/feedback checkpoints for a batch operation, which caps workflow clarity.

Suggestions

Add explicit validation/feedback loops: e.g., 'If search returns <10 relevant videos, broaden keywords and retry' and 'Verify each transcript was retrieved before synthesis; fall back to metadata per step 3.'

Trim marketing language ('absolute truth', 'save the user from hour-long research', 'knowledge artifact that does not exist elsewhere') to reduce token overhead without losing the orchestrator framing.

Add a short output-validation step confirming every major claim has at least one ?t= jump-link before delivering the briefing, closing the batch-operation loop.

Dimension	Reasoning	Score
Conciseness	The body is direct and command-like with no padding of concepts Claude already knows; the only over-explanation is marketing-flavored phrasing ('absolute truth', 'save the user from hour-long research', 'knowledge artifact that does not exist elsewhere') that could be trimmed, fitting the 'efficient; minor instances of over-explanation' anchor.	4 / 5
Actionability	It specifies concrete tool calls with parameters (search_youtube max_results=15, generate_timestamp_url ?t=...) and explicit thresholds (top 10, 3+ creators); as an instruction-only orchestration skill it is mostly executable with minor gaps (no error-path specifics), matching the 'mostly executable guidance' anchor.	4 / 5
Workflow Clarity	The 6-step sequence is clearly ordered, but this is a batch operation over 10+ videos with no validation checkpoints or feedback loops (e.g., verify transcript coverage before synthesizing, retry on empty search), and the judging guidelines cap batch operations lacking validation at 3.	3 / 5
Progressive Disclosure	It is a single, self-contained SKILL.md (no bundle files present) with clear section headers and a well-organized overview; structure is good and appropriately placed, sitting at 'good structure; most content appropriately placed' rather than the fully optimized multi-file anchor.	4 / 5
	Total	15 / 20 Passed

Description

53%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 conveys a specific niche and clear outcomes but omits any 'Use when' trigger guidance and relies on internal jargon over natural user phrases. It is specific enough to be distinct yet incomplete as a discovery trigger.

Suggestions

Add an explicit 'Use when...' clause naming natural triggers, e.g. 'Use when the user wants to synthesize or compare 10+ YouTube videos, find consensus, or map controversies across multiple videos.'

Replace jargon like 'semantic landscapes' and 'targeted extraction' with concrete user-facing actions ('summarize', 'compare', 'cross-reference transcripts').

List concrete actions explicitly (search, filter, transcribe, cross-reference, generate timestamped report) to lift specificity toward a comprehensive action list.

Dimension	Reasoning	Score
Specificity	The description names the domain ('extraction across 10+ videos') and maps 2-3 concrete outcomes ('semantic landscapes, consensus, and controversies'), but 'high-density targeted extraction' is abstract rather than a concrete action list, so it sits at the 'names domain and 1-2 concrete actions' anchor rather than the comprehensive level above.	3 / 5
Completeness	The 'what' is clear (extraction and mapping of consensus/controversies), but there is no 'Use when...' clause or equivalent trigger guidance, which per the judging guidelines caps completeness at 3.	3 / 5
Trigger Term Quality	'videos' is a natural term, but 'semantic landscapes', 'targeted extraction', and 'consensus/controversies' are internal jargon rather than phrases users naturally say; common variations like 'YouTube', 'summarize', or 'compare videos' are missing, matching the 'some relevant keywords but missing common variations' anchor.	3 / 5
Distinctiveness Conflict Risk	The '10+ videos' scope plus 'semantic landscapes/consensus/controversies' niche is mostly distinct with only minor overlap risk against general video-summary skills, fitting 'mostly distinct; minor overlap risk' better than the fully distinct anchor.	4 / 5
	Total	13 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: google/adk-samples
Path: python/agents/youtube-analyst/youtube_analyst/skills/multi-video-synthesis/SKILL.md
Commit: 2d0c272

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.