The Agent Skills Directory

[Prompt Injection] (SAFE): No instructions designed to bypass safety filters or override agent behavior were detected. The skill uses standard instructional language for its intended purpose.
[Data Exposure & Exfiltration] (SAFE): The skill does not access sensitive file paths (e.g., credentials, SSH keys) or perform any network operations to send data externally.
[Unverifiable Dependencies & RCE] (SAFE): No external packages or remote scripts are downloaded or executed. The skill is entirely contained within the markdown file.
[Indirect Prompt Injection] (LOW):
Ingestion points: User responses provided during the multi-round interview process (SKILL.md).
Boundary markers: Absent; the skill interpolates user insights directly into a draft structure.
Capability inventory: The skill is limited to text generation and structured question tool (AskUserQuestion). It cannot execute shell commands or perform network requests.
Sanitization: No specific sanitization or filtering of user-provided insights is described before they are included in the draft output.

blog-drafter