Security audit
Security checks for vulnerabilities and agentic risk
This skill is a disclosed WeryAI video-generation helper for hair makeover clips, with expected use of a WeryAI API key and network calls but no hidden persistence, local data harvesting, or unrelated privileged behavior found.
Install only if you are comfortable giving the skill a WeryAI API key and sending prompts or public image URLs to WeryAI. Review the confirmation table before approving a generation because successful runs consume credits, and watch for the audio setting until the documentation inconsistency is fixed.
The published description frames the skill as narrowly for hair makeover videos, but the documented behavior and referenced script support general-purpose video generation, status polling, model listing, and non-vertical / arbitrary-subject generation. This mismatch can mislead users and reviewers into granting trust, permissions, and secrets under a narrower threat model than the skill actually operates under.
Referenced artifact was not completely inspected
**Dependencies**: `scripts/video_gen.js` in this directory + `WERYAI_API_KEY` + Node.js 18+. No dependency on other Cursor skills.
Referenced artifact was not completely inspected
**Dependencies**: `scripts/video_gen.js` in this directory + `WERYAI_API_KEY` + Node.js 18+. No dependency on other Cursor skills.
Referenced artifact was not completely inspected
**Dependencies**: `scripts/video_gen.js` in this directory + `WERYAI_API_KEY` + Node.js 18+. No dependency on other Cursor skills.
Referenced artifact was not completely inspected
**Dependencies**: `scripts/video_gen.js` in this directory + `WERYAI_API_KEY` + Node.js 18+. No dependency on other Cursor skills.
Referenced artifact was not completely inspected
**Dependencies**: `scripts/video_gen.js` in this directory + `WERYAI_API_KEY` + Node.js 18+. No dependency on other Cursor skills.
Referenced artifact was not completely inspected
- Does not hard-code absolute paths in the skill doc; `{baseDir}` means the skill package root (same level as `SKILL.md`).
The skill requires network access and reads a secret from the environment, but it does not declare an explicit tool scope such as allowed tools or permissions. That weakens sandboxing and reviewability, making it easier for the skill or its companion script to gain broader capabilities than intended and harder for operators to enforce least privilege.
The implicit trigger prompt is broad and style-based rather than tightly scoped to the skill’s intended capability, so unrelated salon, fashion, pet-grooming, or generic makeover requests could spuriously invoke this skill. In an agent system, ambiguous triggering can route user input to the wrong tool, causing unintended external actions, incorrect outputs, or bypass of more appropriate safeguards for adjacent domains.
The description says to use the skill whenever a user needs a hair transformation reel or asks for a related clip, but it does not define specific trigger phrases, exclusions, or narrow activation constraints. This makes invocation scope somewhat broad for a manifest/markdown context and could cause the skill to match loosely related everyday requests about hair or transformations.
The surrounding documentation and parameter table state generate_audio should be false, but the concrete command example at L134 submits "generate_audio":true. This is an active contradiction in the skill's own instructions about what should be sent to the API.
No suspicious patterns detected.