production-optimizer

Pass

Audited by Gen Agent Trust Hub on Mar 23, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes content from untrusted manufacturing codebases without implementing boundary markers or instructions to ignore embedded commands.\n
  • Ingestion points: The skill is directed in Phase 1 and Phase 2 to read various project files, including package managers and source code, to map the production architecture.\n
  • Boundary markers: The instructions lack delimiters or explicit directives to the agent to ignore any natural language instructions that might be contained within the audited files.\n
  • Capability inventory: The skill is designed for the CLAUDE_CODE platform, which provides the agent with filesystem access and command execution capabilities that could be abused if an injection is successful.\n
  • Sanitization: No sanitization, validation, or filtering of the ingested codebase content is specified before the data is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 23, 2026, 10:57 AM
Security Audit — agent-trust-hub — production-optimizer