resemble-detect

Deepfake detection and media safety — detect AI-generated audio, images, video, and text, trace synthesis sources, apply watermarks, verify speaker identity, and analyze media intelligence using Resemble AI

Quality

72%

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 ./resemble-detect/SKILL.md

Quality

Content

77%

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 API reference with strong workflow guidance and explicit checkpoints. Its main weakness is monolithic structure: detailed per-capability reference lives entirely inline with no bundle files to split it out.

Suggestions

Move per-capability detailed reference (Phase response schemas, full parameter tables) into reference files (e.g., references/detect-api.md, references/intelligence-api.md) and link to them one level deep from SKILL.md.

Tighten inline verbosity by condensing full JSON response examples into the essential fields, or relocate full schemas to the reference files.

Keep SKILL.md as an overview with the capability decision tree, core workflow, and red flags, pointing out to the detailed references for each phase.

Dimension	Reasoning	Score
Conciseness	The body is dense reference material without patronizing concept explanations, but full response JSON blocks, multiple workflow variants, and a 'Response Presentation Guidelines' section could be tightened or externalized.	2 / 3
Actionability	Provides concrete endpoints, full request bodies, parameter tables with required flags, polling commands, and explained response fields — copy-paste ready and executable.	3 / 3
Workflow Clarity	Multi-step flows are sequenced (submit → poll → read results) with explicit polling backoff, status-flow transitions, a 422 prerequisite check, an error-handling table, and a 'Red Flags' checklist acting as validation checkpoints.	3 / 3
Progressive Disclosure	No references/, scripts/, or assets/ bundles exist and all per-capability API reference (parameter tables, response schemas, six phases) is inline in a single ~558-line file rather than split into one-level-deep referenced files.	2 / 3
	Total	10 / 12 Passed

Description

67%

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 specific, capability-rich description anchored to a clear platform niche, but it lacks an explicit 'Use when' trigger clause and is light on the natural-language terms users would actually say. Adding trigger phrasing and common user variations would raise completeness and trigger-term quality.

Suggestions

Append an explicit 'Use when...' clause naming concrete trigger moments (e.g., 'Use when checking if audio, video, image, or text is AI-generated or deepfaked').

Add natural user-facing terms to the keyword coverage such as 'is this real', 'fake detection', 'voice verification', and 'AI-written text'.

Consider trimming the abstract 'media safety' framing in favor of the more distinctive 'deepfake detection' wording to sharpen distinctiveness.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions across media types — 'detect AI-generated audio, images, video, and text, trace synthesis sources, apply watermarks, verify speaker identity' — matching the 'multiple specific concrete actions' anchor.	3 / 3
Completeness	Clearly states what the skill does, but there is no 'Use when...' clause or equivalent explicit trigger guidance, which per the guidelines caps completeness at 2.	2 / 3
Trigger Term Quality	Has genuine natural terms ('deepfake', 'AI-generated') but leans technical ('synthesis sources', 'media intelligence') and omits common user variations like 'is this real', 'fake detection', or 'AI-written text'.	2 / 3
Distinctiveness Conflict Risk	Tied to a named platform (Resemble AI) and a specific niche (deepfake detection / media forensics), giving it a clear niche with distinct triggers unlikely to conflict with unrelated skills.	3 / 3
	Total	10 / 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
skill_md_line_count	SKILL.md is long (558 lines); consider splitting into references/ and linking	Warning

	Total	15 / 16 Passed

Repository: Prat011/awesome-llm-skills
Path: resemble-detect/SKILL.md
Commit: 35e1ea2

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.