The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were identified. The skill is entirely composed of markdown documentation and JSON evaluation files without executable code.
[PROMPT_INJECTION]: No instructions were found that attempt to bypass AI safety guidelines or override system prompts. The content is focused on marketing strategy and compliance.
[DATA_EXFILTRATION]: The skill does not perform any network requests or access sensitive local files like SSH keys or environment secrets. All referenced domains are well-known marketing services.
[REMOTE_CODE_EXECUTION]: There are no scripts, command-line executions, or remote download patterns in the skill files. It does not attempt to install or run external software.
[INDIRECT_PROMPT_INJECTION]: The skill includes instructions to read context from files like .agents/product-marketing.md to personalize responses.
Ingestion points: .agents/product-marketing.md, .claude/product-marketing.md, and product-marketing-context.md (referenced in SKILL.md).
Boundary markers: Absent; the skill does not use specific delimiters to separate this context from its primary instructions.
Capability inventory: None; there are no subprocess calls, file-write operations, or network tools available in the skill that could be exploited.
Sanitization: Absent; the skill relies on the agent's default handling of text context.
[OBFUSCATION]: No obfuscated text, encoded URLs, or hidden characters were detected during the analysis of the documentation or templates.

sms