monitor

Pass

Audited by Gen Agent Trust Hub on Oct 4, 2026

Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the Bash tool to execute a local script api-request.sh for interacting with the Spotify Ads API and performing health diagnostics.
  • [DYNAMIC_EXECUTION]: The skill employs eval $(api --env) in the SKILL.md file to dynamically assign environment variables (such as TOKEN and AD_ACCOUNT_ID) based on the output of a local execution wrapper. While this is a common pattern for CLI tools to export configuration, it represents runtime code assembly.
  • [INDIRECT_PROMPT_INJECTION]: The skill fetches external data, including campaign and ad set names, from the Spotify Ads API and renders them directly into markdown tables for the user.
  • Ingestion points: API responses from ad_accounts/{ad_account_id}/campaigns and ad_sets endpoints referenced in SKILL.md.
  • Boundary markers: Data is delimited within markdown table structures.
  • Capability inventory: The skill has access to the Bash tool for further API interactions and the Read tool for accessing local files.
  • Sanitization: No explicit sanitization or escaping of campaign-controlled strings is defined in the prompt instructions before rendering to the agent's context.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 4, 2026, 02:52 AM