writing-for-agents
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill provides conceptual guidelines and structural advice for creating agent instructions (context pointers, cognitive load, information hierarchy). It does not perform any functional tasks that interact with the system or external network.
- [NO_CODE]: The files consist entirely of markdown documentation. No scripts, binaries, or automated tasks are included or referenced.
- [PROMPT_INJECTION]: While the skill discusses prompting techniques, it does not contain any instructions designed to bypass safety filters or override the agent's core system instructions.
Audit Metadata