The Agent Skills Directory

[SAFE]: The skill is well-scoped, limiting its write operations to specific performance testing directories and strictly forbidding modification of application source code or infrastructure configurations.- [EXTERNAL_DOWNLOADS]: The skill references standard installation methods for well-known performance testing tools, including Homebrew for k6 and the official Grafana Docker image (grafana/k6), which are recognized as trusted sources.- [INDIRECT_PROMPT_INJECTION]: The skill processes external API contracts and build plans to generate its testing logic, which is a potential surface for indirect injection. However, this is inherent to the primary function of a performance testing agent and is mitigated by the skill's restricted execution environment.
Ingestion points: api_contract and plan_excerpt files as described in SKILL.md.
Boundary markers: Not explicitly defined in instructions.
Capability inventory: The skill utilizes Bash, Write, and Edit tools to generate and run test scripts.
Sanitization: No explicit sanitization of contract data is mentioned, but behavior is constrained by instructions to follow patterns in references/k6-patterns.md.

performance-agent