The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes shell commands to run test suites and check file differences (e.g., node --test, pnpm test, git diff --check). These are standard operations for a testing and auditing skill and do not involve unauthorized access or dangerous parameters.
[INDIRECT_PROMPT_INJECTION]: The skill's primary function involves reading and analyzing repository source code and test files, creating an ingestion surface for potential indirect prompt injection. 1. Ingestion points: The agent is instructed to read test files and production owner/caller files. 2. Boundary markers: The skill refers to repository test integrity rules (AGENTS.md) as a guide. 3. Capability inventory: Access to the local file system and execution of local test commands. 4. Sanitization: No specific input sanitization is described in the instructions. The overall risk is low as the operations are local and consistent with standard development practices.

test-audit