Skip to content

Add resemble-detect skill to all-skills plugin#123

Open
devshahofficial wants to merge 1 commit intodavepoon:mainfrom
devshahofficial:add-resemble-detect-skill
Open

Add resemble-detect skill to all-skills plugin#123
devshahofficial wants to merge 1 commit intodavepoon:mainfrom
devshahofficial:add-resemble-detect-skill

Conversation

@devshahofficial
Copy link
Copy Markdown

Summary

Adds the resemble-detect skill to the all-skills plugin — a deepfake detection and media safety skill that lets agents detect AI-generated audio, images, video, and text using the Resemble AI platform.

What the skill does

  • Deepfake detection across all 4 modalities (audio, image, video, text)
  • Source tracing — identify which AI platform likely generated a given audio sample
  • Watermarking — apply and detect invisible watermarks for provenance
  • Speaker identity — voice profile matching and verification
  • Media intelligence — emotion, language, abnormality, and voice metadata analysis

Core principle ("Iron Law")

The skill instructs agents to never declare media as real or fake without a completed detection job — every authenticity claim must be backed by a returned label, score, and status: "completed" from the Resemble API. No guessing, no speculation.

Changes

  • New: plugins/all-skills/skills/resemble-detect/SKILL.md — full skill definition with decision tree, API reference, workflow templates, red flags, and error handling
  • New: plugins/all-skills/skills/resemble-detect/LICENSE — Apache-2.0
  • Updated: plugins/all-skills/.claude-plugin/plugin.json
    • Description bumped from 30 to 31 skills; added "media safety" to the topic list
    • Added resemble-detect to keywords

Test plan

  • npm run validate — all validations passed (subagents & hooks)
  • SKILL.md frontmatter includes name, category: analytics, description, license
  • Skill folder follows the documented pattern: plugins/all-skills/skills/<skill-name>/SKILL.md
  • plugin.json keywords array updated

Links

Deepfake detection and media safety skill — detects AI-generated
audio, images, video, and text using Resemble AI. Includes audio
source tracing, watermarking, speaker identity verification, and
media intelligence workflows.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant