The Agent Skills Directory

[Data Exposure & Exfiltration] (SAFE): The skill does not attempt to access sensitive directories such as .ssh or .aws, nor does it perform network operations to non-whitelisted domains. It primarily uses the Write tool to store documentation locally.
[Prompt Injection] (SAFE): No instructions were found that attempt to bypass safety filters, ignore prior rules, or extract system prompts.
[Obfuscation] (SAFE): The skill content is entirely plain-text markdown and YAML. There are no instances of Base64, zero-width characters, or other encoding techniques.
[Unverifiable Dependencies & Remote Code Execution] (SAFE): No external packages (npm/pip) are requested, and no remote scripts are downloaded or executed.
[Indirect Prompt Injection] (SAFE): The skill analyzes local codebase content which constitutes an untrusted data surface, but its capabilities are restricted to documentation generation. 1. Ingestion points: Reads codebase files using Read, Glob, and Grep tools (SKILL.md). 2. Boundary markers: Absent. 3. Capability inventory: Uses Write tool to create documentation and Skill tool to call a diagram generator. 4. Sanitization: Absent, which is expected for static documentation generation.

architecture-documentation