The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute various shell commands for build verification (npm run build, npx tsc), testing (npm run test), and security auditing (npm audit).
[COMMAND_EXECUTION]: Interacts with external CLI tools and infrastructure including the AWS CLI (aws configure get region) and Vercel CLI (npx vercel inspect).
[INDIRECT_PROMPT_INJECTION]: Ingests and processes data from external sources such as git diff and tool outputs, which could contain malicious instructions embedded by an external actor.
Ingestion points: Raw text from git diff is read and analyzed in Phases 3, 5.1, and 5.2 within SKILL.md.
Boundary markers: Absent; there are no instructions to delimit or ignore potentially malicious content within the processed data.
Capability inventory: The skill has access to the local file system (via git), network operations (via curl), and arbitrary process execution (via npm and npx).
Sanitization: Absent; the agent is not instructed to validate or escape the contents of the git diff or command outputs before processing them.

qa-checklist