The Agent Skills Directory

[SAFE]: A thorough review of all skill files, including the markdown instructions and YAML templates, reveals no evidence of malicious patterns, unauthorized data access, or command injection. The skill operates within its described scope of business analysis and project documentation.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input to generate business documents, creating a surface for potential indirect prompt injection.
Ingestion points: Found in SKILL.md commands (e.g., brainstorm {topic}) and across all template resource files where user-supplied strings are used to fill template variables.
Boundary markers: Uses standard {{variable}} interpolation syntax within YAML template files.
Capability inventory: The skill writes generated content to the filesystem (specifically the docs/ directory) via the create-doc helper skill.
Sanitization: No explicit input sanitization or safety boundary instructions were identified in the static templates. However, this surface is associated with the primary intended purpose of the skill and is categorized as low risk.

analyst