writing-for-agents
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of markdown documentation and YAML metadata. It provides conceptual frameworks for structuring information to be consumed by AI agents.
- [NO_CODE]: The skill does not bundle any scripts, binaries, or automated shell commands. It relies purely on instructional text to guide the agent's behavior during document creation tasks.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to assist in creating and editing agent-consumed documents. While this involves processing potentially untrusted content, the skill explicitly includes safety guidance in its 'Anti-Patterns' section, instructing agents not to treat external content, logs, or tool responses as trusted instructions. This mitigation reduces the risk of the agent following malicious instructions embedded in documents it processes.
Audit Metadata