The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to execute test and build commands resolved from project artifacts (Steps 6–7). This is the intended primary purpose of the verification skill and is performed within the project's local context.
[DATA_EXPOSURE]: The skill reads from and writes to the local filesystem (e.g., docs/sdd/{change-name}/verify-report.md). These operations are scoped to the project directory and used to manage verification results.
[REMOTE_CODE_EXECUTION]: The skill executes a local script node core/hooks/_shared/receipt-mint.js to attest to the review. This script is part of the vendor's internal shared infrastructure and does not involve downloading or executing code from external, untrusted sources.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from project artifacts to determine verification status. While this constitutes an input surface for indirect injection, the skill's capabilities (execution of build/test commands) are consistent with its role as a quality gate tool, and no exploitation patterns were found.

sdd-verify