skills/sickn33/agentic-awesome-skills/performance-testing-review-multi-agent-review/Gen Agent Trust Hub
performance-testing-review-multi-agent-review
Pass
Audited by Gen Agent Trust Hub on Aug 10, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its processing of external code artifacts. 1. Ingestion points: The skill accepts target code snippets and project data via the
$ARGUMENTSparameter (identified in SKILL.md). 2. Boundary markers: There are no explicit delimiters or instructions to the agent to disregard instructions found within the code being analyzed. 3. Capability inventory: The orchestration logic involves routing input to various expert agents (Security, Architecture, etc.) and synthesizing results, a process that could be influenced by malicious comments in the source code. 4. Sanitization: No validation or sanitization is specified for the code artifacts being processed.
Audit Metadata