The Agent Skills Directory

[SAFE]: The skill is purely instructional and advisory. It contains no executable scripts, binaries, or shell commands, and it does not perform any network operations.- [DATA_EXPOSURE_AND_EXFILTRATION]: The skill instructs the agent to read project-specific context from files like .agents/product-marketing.md. This is a standard practice for tailoring responses to a specific project and does not involve accessing sensitive system files (e.g., SSH keys) or credentials.- [INDIRECT_PROMPT_INJECTION]: The skill defines a potential attack surface by ingesting data from external context files, but since the skill lacks any dangerous capabilities, this surface is not exploitable for system harm.
Ingestion points: Reads .agents/product-marketing.md, .claude/product-marketing.md, and product-marketing-context.md to gather business context.
Boundary markers: None present; the skill treats the content of these files as trusted project documentation.
Capability inventory: None; the skill is advisory and does not utilize tools for file system modification, network communication, or code execution.
Sanitization: None; the agent interprets the context files directly as natural language instructions.

lead-magnets