The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to perform operations such as gh pr view, gh pr diff, and potentially build or lint commands suggested in the 'Fitness Functions' and 'PR Review' sections. This is necessary for its intended functionality of code and architectural review.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection during its review processes.
Ingestion points: The skill ingests untrusted data from external sources when performing PR reviews using the GitHub CLI (gh).
Capability inventory: The skill has access to capabilities including Bash, Write, Read, and Glob as defined in its YAML frontmatter.
Boundary markers: There are no explicit instructions or delimiters defined to isolate untrusted PR content from the agent's core instructions.
Sanitization: No sanitization or validation of the PR content (diffs or descriptions) is performed before the agent processes and reasons about the data.

arch