The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from multiple sources and passes them to sub-agents without sufficient isolation or sanitization.
Ingestion points: git diff output, git log output (commit messages), docs/agents/issue-tracker.md, and various repository files including specs, standards, and files under .scratch/.
Boundary markers: The instructions for the sub-agent prompts in Step 4 do not specify the use of delimiters or 'ignore embedded instructions' warnings for the interpolated data.
Capability inventory: The skill utilizes git commands and a subagent tool to process and report findings.
Sanitization: There is no evidence of sanitization, escaping, or filtering of the external content before it is interpolated into the prompts for the sub-agents.
[COMMAND_EXECUTION]: The skill uses local shell commands to facilitate the code review process.
Evidence: It executes git diff, git log, and git rev-parse to analyze the repository state and identify changes since a specified fixed point. These are standard and intended operations for the skill's purpose.

code-review