The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes benign Git commands (git log, git rev-parse) to gather repository information such as commit counts and the absolute path to the repository root. These operations are used solely for project reconnaissance and environment discovery.
[INDIRECT_PROMPT_INJECTION]: The skill reads external codebase files to identify project conventions and 'prior art'. This represents an attack surface for indirect prompt injection, as malicious instructions in codebase comments could attempt to influence the agent's logic. This risk is mitigated by a mandatory manual review gate where the user must approve the generated specification before any permanent files are created.
[DATA_EXPOSURE]: The skill identifies and reads local source files to assist in feature planning. This is the primary function of the skill and is performed within the local execution environment; no evidence was found of data being transmitted to external or untrusted domains.

slice