The Agent Skills Directory

[SAFE]: The skill focuses on repository documentation and marketing strategy. It performs legitimate file read and write operations restricted to the local project environment.\n- [DATA_EXFILTRATION]: The skill reads non-sensitive project files like README, package.json, and marketing assets to assist in drafting content. No credentials, sensitive system files, or external network calls are detected.\n- [REMOTE_CODE_EXECUTION]: There are no patterns of remote code execution, external script fetching, or unauthorized command execution.\n- [PROMPT_INJECTION]: The skill ingests codebase content (README, landing pages) which constitutes a potential attack surface for indirect prompt injection. However, the agent's role is restricted to drafting marketing context, and the risk is minimal for this specific use case. Ingestion points: README, landing pages, package.json (SKILL.md). Boundary markers: None. Capability inventory: file read/write. Sanitization: None.

product-marketing