Skills
Navigate
skills/kevanstannard-atomi/agent-skills/atomi-staging-db/Gen Agent Trust Hub

atomi-staging-db

Pass

Audited by Gen Agent Trust Hub on Mar 30, 2026

Risk Level: SAFEDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [DATA_EXFILTRATION]: The file references/account-users.md contains the names and email addresses of internal staff members (e.g., Andrew McLagan, Yury B., Atomi Admin) associated with staging test accounts. While intended for a testing environment, this constitutes exposure of internal personnel data.
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it is designed to ingest and process data from external MySQL databases.
  • Ingestion points: Data retrieved via the dbhub MCP server from the content_stage and nucleus_stage databases, including user-generated content fields in the QuestionItem, Question, posts, and TextSnippet tables.
  • Boundary markers: The skill instructions do not establish delimiters or provide "ignore embedded instructions" warnings for the content retrieved from the database queries.
  • Capability inventory: The skill is configured to execute SQL queries using the dbhub MCP server to interact with staging data.
  • Sanitization: There is no evidence of content sanitization or validation for the data retrieved from the database before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 30, 2026, 12:03 PM