The Agent Skills Directory

[SAFE]: The skill consists entirely of markdown instructions and templates, with no scripts or executables provided. All logic is handled via the agent's natural language processing.
[DATA_EXPOSURE]: The instructions include proactive safety measures, such as explicit warnings against requesting sensitive credentials or confidential data and guidance on redacting information for diligence preparation.
[INDIRECT_PROMPT_INJECTION]: The skill represents a safe implementation because it lacks access to any dangerous capabilities. Ingestion points: User-provided company snapshot and traction data in SKILL.md. Boundary markers: Explicit instructions to avoid sensitive credentials in SKILL.md. Capability inventory: No tools or execution capabilities identified across all skill files. Sanitization: Not specified but unnecessary given the lack of executable tools.

fundraising