realtime-control-design
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill consists entirely of markdown documentation and design patterns. There are no scripts, shell commands, or automation components provided.
- [SAFE]: All external references target official documentation for the ESP-IDF framework on the
docs.espressif.comdomain, which is a well-known technology service. - [INDIRECT_PROMPT_INJECTION]: The skill's workflow requires the agent to read external project documents and research notes (e.g.,
arduino-skill-contract.md,bottlesumo-lessons.md). While this constitutes an ingestion of untrusted data, the skill possesses no capabilities—such as file system writes, network exfiltration, or command execution—that could be exploited via embedded instructions.
Audit Metadata