The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted external data, which creates a potential vector for indirect prompt injection.
Ingestion points: The agent reads code changes via git diff and pull request comments via the .agents/skills/fix-pr-comments/pr-threads.sh script.
Boundary markers: There are no instructions provided to the agent to use delimiters or ignore instructions that may be embedded within the code diffs or PR comments.
Capability inventory: The skill has the ability to execute shell commands (git), run local shell scripts, and use a python-playground for code execution.
Sanitization: The instructions do not specify any methods for sanitizing, escaping, or filtering the content retrieved from the codebase or pull request threads.
[COMMAND_EXECUTION]: The skill explicitly instructs the agent to execute the git diff command to identify changes between the current branch and the merge base.
[REMOTE_CODE_EXECUTION]: The skill requires the execution of a local shell script located at .agents/skills/fix-pr-comments/pr-threads.sh. While this is a local file, it represents a dependency on external script execution to perform its primary function.

review-security