The Agent Skills Directory

[PROMPT_INJECTION]: The skill presents an indirect prompt injection surface by ingesting and processing external data from task documentation and source code files.
Ingestion points: Reads content from docs/task/{ID}-{task-name}.md and files identified by the git diff command.
Boundary markers: No specific delimiters or "ignore embedded instructions" warnings are utilized when reading file contents.
Capability inventory: Includes shell command execution (git diff), file write access to update task documentation, and the ability to spawn subagents using the Task tool.
Sanitization: No explicit sanitization or validation of the ingested file content is performed prior to analysis.
[COMMAND_EXECUTION]: The skill uses local shell commands to identify the scope of work for its review process.
Evidence: Executes git diff --name-only main...HEAD to programmatically determine which files have changed in the current branch.

simplify