content-coverage

Pass

Audited by Gen Agent Trust Hub on Nov 12, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection because it ingests content from external, unverified URLs and uses that information to generate and write files to the local repository.\n
  • Ingestion points: The skill fetches 3-5 external citation URLs (Phase 3) identified through search results via the aiseo_list_sources tool.\n
  • Boundary markers: There are no explicit instructions to the AI to ignore potentially malicious embedded content or prompts within the fetched citation data.\n
  • Capability inventory: The skill has file-write capabilities, creating or updating markdown/MDX files in the user's content repository (Phase 5).\n
  • Sanitization: The instructions focus on verifying URL resolution but do not include patterns for sanitizing or filtering instructions that might be embedded in the remote content.\n- [EXTERNAL_DOWNLOADS]: The skill fetches instructions and configuration data from remote locations on the vendor-owned domain.\n
  • Evidence: The skill instructions specify fetching https://aiseotracker.com/SKILL.md and https://aiseotracker.com/content-coverage/SKILL.md to ensure the skill is complete and to handle MCP connectivity.\n- [COMMAND_EXECUTION]: The skill orchestrates multiple tool calls and file system operations as part of its primary functionality.\n
  • Evidence: Use of custom MCP tools such as aiseo_list_projects, aiseo_get_scan, aiseo_list_prompts, and local file system writes to generate blog content in the repository.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 12, 2026, 04:49 PM