The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local shell script located at .agents/skills/codex-fix-prose/scripts/check-suppressions.sh using the bash interpreter. This script is used to diff and verify the state of a prose fix. This is a standard functional component of the skill's workflow.
[INDIRECT_PROMPT_INJECTION]: The skill reads and reviews external content (the 'target' file) which may be controlled by untrusted parties. This creates a surface for indirect prompt injection where instructions hidden within the prose could attempt to influence the agent's behavior.
Ingestion points: The skill reads the target file and the workflow state provided by the caller.
Boundary markers: There are no explicit delimiters or instructions to the agent to disregard potential commands found within the processed text.
Capability inventory: The skill has the capability to execute local bash scripts and perform file system reads.
Sanitization: No sanitization or validation of the input content is described before the agent processes the text for review.

codex-review-prose-fix