The Agent Skills Directory

[SAFE]: The skill operates as a structured logging tool, utilizing the Read, Write, and Glob tools to maintain a transmittal history in a specific project directory. It does not perform network operations, access sensitive system files, or execute external code.
[PROMPT_INJECTION]: The skill accepts user-provided text (recipient, purpose, file descriptions) to populate transmittal logs. This creates a surface for indirect prompt injection, as untrusted data is written to the project's file system. However, this functionality is intrinsic to the skill's purpose as a logger, and the use of a rigid Markdown template mitigates the risk of the agent misinterpreting the generated content.
Ingestion points: User input for transmittal metadata and file lists.
Boundary markers: Data is interpolated into a predefined Markdown table and list format.
Capability inventory: The skill uses Write to create log files and Glob to determine numbering.
Sanitization: The instructions require using 'not recorded' for missing information, though they do not specify technical escaping for the user-provided strings.

transmittal