The Agent Skills Directory

[COMMAND_EXECUTION]: The skill automates complex workflows by executing the coval CLI and standard shell utilities like jq and bash. These commands are used to list agents, manage test sets, and launch simulations within the user's own environment.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from its own command outputs to determine configuration for subsequent steps. It uses structured JSON parsing via jq to extract identifiers for agents and personas, which are then passed to execution commands.
Ingestion points: Data extracted from coval agents list, coval test-sets list, and coval personas list as seen in SKILL.md.
Boundary markers: None explicitly defined in the instructions, though reliance on structured CLI output minimizes accidental interpretation.
Capability inventory: Shell command execution via the coval CLI across multiple steps in SKILL.md.
Sanitization: The use of jq ensures data is extracted correctly from structured formats, though the skill does not perform further validation of the content of the IDs themselves.
[EXTERNAL_DOWNLOADS]: The skill documentation provides instructions for installing the official Coval CLI tool and references vendor-hosted documentation and repositories. These downloads are performed from trusted vendor domains and official distribution channels.

run-audio-quality-testing