The Agent Skills Directory

[SAFE]: The skill operates within its intended scope of creating technical specifications and formal models.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection attack surface as it processes untrusted external data from the local repository.
Ingestion points: Reads files from the docs/requirements/ directory and existing Architecture Decision Records (ADRs) to gather context.
Boundary markers: No specific delimiters or markers are defined to isolate ingested file content from the agent's core instructions.
Capability inventory: The skill uses Read, Write, Glob, Grep, Skill, and Task tools, enabling it to search the filesystem and execute complex orchestrated tasks based on its findings.
Sanitization: There is no evidence of sanitization or structural validation for the data ingested from the requirements documents before it is used to generate specifications.

specify