principle-redesign-from-first-principles
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to "Read all affected files" to understand current designs. This establishes a data ingestion surface where the agent processes external, potentially untrusted content from the local codebase. Malicious instructions hidden in those files could theoretically influence the agent's logic during the redesign process.
- [NO_CODE]: The skill consists entirely of natural language instructions within a markdown file. It does not provide any scripts, executable binaries, or configuration files that could carry active payloads.
Audit Metadata