performance-reliability-tester
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process potentially untrusted external data, such as repository documentation, issue text, and requirements, creating an attack surface for indirect prompt injection findings.\n
- Ingestion points: The agent is instructed to read requirements, READMEs, comments, and issue text as part of its testing intake (
references/agent-qa-mechanisms.md).\n - Boundary markers: The skill contains explicit defensive instructions to treat these inputs as data rather than commands and to disregard any embedded instructions that attempt to bypass safety policies (
references/agent-qa-mechanisms.md).\n - Capability inventory: The agent is expected to execute commands and interact with test environments, which could be abused if boundary instructions are ignored (
SKILL.md).\n - Sanitization: Safety relies on prompt-level constraints and explicit "default-deny" rules for destructive actions and credential harvesting.
Audit Metadata