adobe-deploy-integration
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill integrates with official and well-known command-line interfaces, including Adobe I/O (
aio), Vercel (vercel), and Google Cloud (gcloud), for its deployment workflows. - [SAFE]: Credential management instructions follow industry best practices by utilizing platform-native secret management tools (e.g.,
gcloud secrets,vercel env) to inject API keys and secrets as environment variables, avoiding insecure hardcoding. - [INDIRECT_PROMPT_INJECTION]:
- Ingestion points: The Vercel serverless function example in
SKILL.mdingests external data directly fromreq.bodyto pass into a Firefly API request. - Boundary markers: None present in the example code snippets.
- Capability inventory: The skill uses
Bashtools for deployment andfetchfor network operations to Adobe's API. - Sanitization: The provided code is a simplified implementation example and does not include input sanitization, which is a standard surface for indirect prompt injection that developers should address in production implementations.
Audit Metadata