The Agent Skills Directory

[SAFE]: The instructions and referenced files focus exclusively on business growth strategies, frameworks like AARRR and ICE, and product-market fit analysis. No malicious instructions, command execution, or network exfiltration patterns were identified.
[SAFE]: File access is restricted to relative paths within the workspace for context reading and draft saving, adhering to safe data handling practices for agent skills.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it ingests data from external workspace files, though its restricted capabilities minimize the potential risk.
Ingestion points: The skill reads from strategy/brand.md, about/me.md, content/ideas.md, and content/calendar.md to establish project context.
Boundary markers: The skill's 'Operating Contract' provides instructions for scope limitation, but no technical delimiters are implemented for the ingested markdown content.
Capability inventory: Capabilities are limited to reading local files and writing content drafts. No network operations, system command execution, or privileged access patterns were found.
Sanitization: No explicit sanitization or validation of content read from the workspace files is performed before being used in draft generation.

growth-strategy