db-orchestrator

Pass

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local JavaScript files via the Bash tool to perform core functions like stack detection and scoring. Evidence: The execution of scripts/detect-stack.mjs in Phase 1 and scripts/score.mjs in Phase 3.
  • [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection (Category 8) because it reads and processes untrusted data from the user's environment to generate audit reports.
  • Ingestion points: Ingests data from project directories, including source code, DDL, and queries, as well as plain-language descriptions provided by the user.
  • Boundary markers: Absent; the skill does not specify the use of delimiters or 'ignore' instructions when reading content from external files.
  • Capability inventory: The orchestrator and its subagents have access to high-privilege tools such as Bash, Task, and WebFetch.
  • Sanitization: There is no mention of sanitizing or validating the content of the files being read before the agent processes them, which could allow instructions hidden in database schemas to influence agent behavior.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 17, 2026, 10:01 PM
Security Audit — agent-trust-hub — db-orchestrator