qe-iterative-loop

Pass

Audited by Gen Agent Trust Hub on Mar 30, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute a wide range of standard development tools including npm test, npm run lint, npm run typecheck, npm audit, and npx tsc. These commands are used to evaluate and improve code quality through autonomous iteration loops.
  • [SAFE]: The skill references an external GitHub repository (github.com/proffesor-for-testing/agentic-qe) which belongs to the skill author and is considered a vendor-owned resource.
  • [SAFE]: No obfuscation, hardcoded credentials, or unauthorized network operations were detected in the instructions or metadata.
  • [PROMPT_INJECTION]: The skill creates an attack surface for indirect prompt injection by processing untrusted data from the local environment and external sources.
  • Ingestion points: The skill parses output from test runners (e.g., npm test), coverage reports (e.g., c8, istanbul), and quality gate check results.
  • Boundary markers: No explicit boundary markers or instructions to disregard embedded commands within the processed tool outputs are defined.
  • Capability inventory: The agent can execute shell commands (npm, npx), manage tasks via MCP tools (mcp__agentic-qe__task_orchestrate), and access persistent memory.
  • Sanitization: There is no evidence of sanitization, escaping, or validation of the external tool output before it is used to influence the iteration loop's decision-making logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 30, 2026, 02:28 AM