vercel-deploy

Pass

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: SAFEDATA_EXFILTRATIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [DATA_EXFILTRATION]: The skill packages the contents of a local directory into a compressed tarball and transmits it to an external API endpoint via a POST request. Evidence: scripts/deploy.sh uses curl to send data to https://claude-skills-deploy.vercel.com/api/deploy.
  • [COMMAND_EXECUTION]: The deployment script executes multiple system commands to process files and prepare the deployment package. Evidence: Use of tar, find, mv, and mktemp within scripts/deploy.sh.
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection because it ingests a directory path and performs actions with significant capabilities. 1. Ingestion points: The INPUT_PATH argument in scripts/deploy.sh derived from command-line arguments. 2. Boundary markers: Absent; there are no delimiters or warnings to ignore instructions within the processed data. 3. Capability inventory: The skill has the ability to read files (tar), list directory contents (find), and transmit data externally (curl). 4. Sanitization: No validation is performed on the input path to prevent access to sensitive system directories (e.g., SSH keys or configuration files).
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 6, 2026, 03:03 AM
Security Audit — agent-trust-hub — vercel-deploy