The Agent Skills Directory

[COMMAND_EXECUTION]: The skill contains instructions for the AI agent to execute shell commands (ls) to inspect the contents of hidden directories such as ~/.claude/skills/ and ~/.agent/skills/. This is designed to discover companion skills but results in scanning the local filesystem for information outside the immediate working directory.
[EXTERNAL_DOWNLOADS]: The documentation references an external installation command (npx skills add AbsolutelySkilled/AbsolutelySkilled) that downloads and executes code from a remote source.
[PROMPT_INJECTION]: The skill defines patterns for processing and asserting on external API responses, creating a vulnerability surface for indirect prompt injection.
Ingestion points: Data enters via API response bodies (e.g., res.body in supertest) and GraphQL operation results.
Boundary markers: The provided code examples do not include delimiters or instructions to ignore embedded commands within the processed data.
Capability inventory: The agent possesses capabilities for shell command execution, filesystem access, and network operations.
Sanitization: No sanitization or filtering of the incoming data is implemented in the provided test patterns.

api-testing