The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill contains a vulnerability surface for indirect prompt injection as it is designed to ingest and process untrusted project briefs, audience descriptions, and tone requirements.
Ingestion points: The agent reads the concept, purpose, audience, and tone from the provided naming brief as described in SKILL.md.
Boundary markers: The instructions lack explicit delimiters or "ignore embedded instructions" warnings for the ingested brief data.
Capability inventory: The skill's capabilities are limited to text generation, name analysis, and the use of the define-art-direction tool; no file writes, network exfiltration, or shell execution capabilities are present in the skill files.
Sanitization: No sanitization or validation of the brief content is performed before interpolation into the agent's context.
[SAFE]: No high-risk patterns such as remote code execution, credential exfiltration, or obfuscation were detected. The skill's operations are confined to creative text generation and analytical tasks within the agent's safe environment.

brainstorm-names