The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to perform version control operations directly on the local repository, specifically using git add, git commit, and git check-ignore to persist the generated specifications and research files.
[INDIRECT_PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection as it is designed to ingest and synthesize external documentation such as vendor guides, API references, and user-provided conversation threads into formal project specifications.
Ingestion points: Conversation context, external reference links (vendor guides, API docs), and pasted textual source material.
Boundary markers: There are no explicit instructions or delimiters provided to the agent to prevent it from following commands potentially embedded within these external source materials during synthesis.
Capability inventory: The skill allows the agent to read and write to the filesystem (specifically the .scratch/ directory) and execute shell-based git commands.
Sanitization: The process lacks specified mechanisms for sanitizing, escaping, or filtering external content before it is processed or stored beside the specification files.

to-spec