The Agent Skills Directory

[SAFE]: The skill does not contain any malicious code, obfuscation, or unauthorized network operations. It functions as an instructional framework for generating professional documents.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data including entity details, project goals, and financial records. While this represents a potential injection surface, the workflow includes explicit quality gates such as anti-ai-slop and meta-critical-thinking-business-logic to sanitize and validate outputs. Additionally, it mandates parameterization of client inputs to ensure they are reviewed rather than executed as facts.
Ingestion points: Entity types, programme goals, theory of change, funding models, and existing frameworks (SKILL.md).
Boundary markers: Not explicitly defined in the instructions, but parameterization rules act as a functional boundary.
Capability inventory: None; the skill is primarily focused on text generation and document assembly.
Sanitization: Implementation of anti-ai-slop and critical-thinking gates in the workflow to review logic and consistency.

me-framework-document