skills/ronaldocutrim/skills/ai-sdk/Gen Agent Trust Hub

ai-sdk

Pass

Audited by Gen Agent Trust Hub on May 7, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill implements an indirect prompt injection surface by fetching and processing external documentation content. Ingestion points: Fetches content from ai-sdk.dev. Boundary markers: Absent; the agent is not instructed to use specific delimiters or ignore instructions in fetched text. Capability inventory: Executes local search commands (grep, glob) and reads files. Sanitization: Absent; content is processed directly to answer queries. Additionally, a discrepancy exists between the platform author (ronaldocutrim) and the author field in the skill metadata (Vercel Inc.), which is misleading.
  • [EXTERNAL_DOWNLOADS]: Fetches search queries and markdown documentation from the official Vercel AI SDK documentation domain.
  • [COMMAND_EXECUTION]: Uses grep and glob commands to search for documentation and source code within the local node_modules/ai/ and @ai-sdk/ provider directories.
Audit Metadata
Risk Level
SAFE
Analyzed
May 7, 2026, 08:08 AM