The Agent Skills Directory

[PROMPT_INJECTION]: The skill creates an indirect prompt injection surface by processing untrusted data from the repository and incorporating it into instructions for autonomous sub-agents.
Ingestion points: The skill reads input from git diff, git log (including commit messages), and external specification documents (Issues and PRDs).
Boundary markers: The prompt templates for the sub-agents do not include delimiters or instructions to disregard embedded commands within the analyzed content.
Capability inventory: The skill uses the Agent tool to execute general-purpose analysis on the ingested data.
Sanitization: No validation or sanitization is performed on the text extracted from the git history or external specifications before it is used to construct sub-agent prompts.
[COMMAND_EXECUTION]: The skill relies on local shell commands to retrieve repository data and verify references.
Evidence: It executes git diff, git log, and git rev-parse using parameters derived from user input.

code-review