The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external, untrusted content from git diffs and specification files and interpolates this data into the prompts for sub-agents without strict boundary markers or sanitization.
Ingestion points: git diff output, git log output, and content from files in docs/, specs/, or .scratch/.
Boundary markers: The prompt templates for the Standards and Spec sub-agents do not include explicit delimiters or instructions to ignore embedded commands within the diffs/specs.
Capability inventory: The skill utilizes the Agent tool to spawn sub-agents and executes multiple git shell commands.
Sanitization: There is no evidence of filtering or sanitization of the external content before it is processed by the LLM.
[COMMAND_EXECUTION]: The skill executes various git commands (git diff, git log, git rev-parse) using user-provided references. While it includes a validation step (git rev-parse), there is a minor risk of command injection if the underlying execution environment does not properly escape arguments.

code-review