The Agent Skills Directory

[SAFE]: The skill uses Bash tools for structural project analysis, such as identifying source directories and file counts. These actions are performed locally within the project scope to inform documentation generation and do not involve data exfiltration or unauthorized system access.
[PROMPT_INJECTION]: The skill presents an indirect prompt injection surface, as it ingests project file content and directory metadata to generate development context.
Ingestion points: Reads project-specific data using Read, Glob, and Bash tools.
Boundary markers: None; instructions do not use delimiters to isolate ingested code from the agent's reasoning process.
Capability inventory: The skill uses Write, Edit, and Task (agent delegation) tools to create files and manage sub-agents based on ingested data.
Sanitization: There is no explicit sanitization of the project content before it is processed by the agent.

update-claudes