skills/diegosouzapw/awesome-omni-skills/application-performance-performance-optimization-v2/Gen Agent Trust Hub
application-performance-performance-optimization-v2
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data via the
$ARGUMENTSplaceholder and interpolates it directly into prompts for specialized sub-agents (e.g., performance-engineer, database-optimizer, cloud-architect). This creates a vulnerability where malicious input could influence or override the sub-agent's instructions. - Ingestion points: User-supplied input enters the workflow through the
$ARGUMENTSvariable used in nearly all phase-specific instructions inSKILL.md. - Boundary markers: The skill fails to define explicit delimiters or use "ignore embedded instructions" warnings when passing user data to sub-agents.
- Capability inventory: The sub-agents invoked by this skill have extensive capabilities, including profiling production systems, optimizing database queries/indexes, modifying backend code, and executing load testing tools (k6, Artillery, Gatling).
- Sanitization: There is no evidence of validation, escaping, or filtering of the user-provided arguments before they are processed by the sub-agents.
Audit Metadata