The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data in the form of git diffs and source code files for analysis.
Ingestion points: The workflow in SKILL.md and subagent instructions in references/bug-security-reviewer.md collect git diffs and changed file contents.
Boundary markers: The subagents are instructed to identify data within labeled sections such as ### Git / diff output and ### Changed file contents.
Capability inventory: The skill utilizes git for diffing, gh or glab for reading pull request discussions, and read-only file system access.
Sanitization: No explicit sanitization or filtering of the ingested code content is implemented to prevent potential instruction injection within the code being reviewed.
[COMMAND_EXECUTION]: The instructions involve the use of standard development command-line tools for context gathering.
Evidence: The agent is directed to use git, gh (GitHub CLI), and glab (GitLab CLI) to gather diffs and discussions related to code changes. These tools are used for their intended purpose of retrieving project-related metadata and code state.

cursor-thermos