The Agent Skills Directory

[SAFE]: The skill instructions are purely procedural and focus on summarizing existing context into a document format. It does not introduce any shell commands, external downloads, or persistence mechanisms.
[PROMPT_INJECTION]: The skill processes potentially untrusted data from the conversation history and repository context to populate its specification template. Ingestion points: Current conversation context and codebase understanding. Boundary markers: Employs a XML-like tag to delineate the generated document. Capability inventory: Uses the agent's capability to write content to a project issue tracker. Sanitization: No explicit escaping or sanitization of input data is defined, relying on the agent's synthesis logic.

to-spec