The Agent Skills Directory

[DATA_EXPOSURE_AND_EXFILTRATION]: The skill instructions suggest checking project-specific files like .claude/project-context.md or .cursor/project-context.md for keyword and strategy context. This is a standard practice for project-aware agents and does not involve accessing sensitive system-level credentials or private user data.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it ingests data from project context files.
Ingestion points: .claude/project-context.md, .cursor/project-context.md (SKILL.md).
Boundary markers: None explicitly defined for these files.
Capability inventory: The skill only generates text and structural recommendations for Medium articles; it does not have file-write, network-send, or code-execution capabilities.
Sanitization: None detected.
[SAFE]: No malicious patterns, obfuscation, or remote code execution vectors were detected in the skill content. It functions as a structured guide for content distribution best practices.

medium-posts