The Agent Skills Directory

[NO_CODE]: The skill is comprised entirely of Markdown-based instructions, tables, and assessment frameworks. It does not include any scripts (Python, Node.js, Shell), binaries, or configuration files that execute code.
[INDIRECT_PROMPT_INJECTION]: The skill defines clear ingestion points for untrusted data provided by the user (e.g., organizational pain points, stakeholder landscape, and strategic context). While it lacks explicit delimiters or sanitization for these inputs, the skill does not possess any exploitable capabilities, such as network access, file system modification, or command execution, that would allow for an injection attack to cause harm beyond influencing the generated report.
Ingestion points: 'Required Inputs' section in SKILL.md.
Boundary markers: Absent.
Capability inventory: None. The skill is purely analytical and produces text-based output without calling external tools or scripts.
Sanitization: Absent.
[SAFE]: The content was thoroughly scanned for obfuscation (Base64, zero-width characters, homoglyphs), persistence mechanisms, and privilege escalation patterns. None were found. The skill operates as a static template for organizational consulting.

digital-transformation