pci-dss-compliance
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external artifacts, creating an indirect prompt injection surface.
- Ingestion points: The agent reads repositories, diffs, and various compliance documents (e.g., SAQs, policies) during the review process.
- Boundary markers: Absent; there are no instructions to use delimiters or ignore instructions within the ingested artifacts.
- Capability inventory: The skill utilizes shell-based search tools (ripgrep) and file system write operations to save reports.
- Sanitization: Absent; the skill does not instruct the agent to escape or sanitize external content before processing it.
- [COMMAND_EXECUTION]: The skill provides predefined shell commands using
ripgrep(rg) to identify cardholder data and payment SDKs in the codebase (references/code-review-checklist.md). This is a legitimate audit function. - [DATA_EXFILTRATION]: The skill searches for sensitive account data but includes explicit instructions to mask PANs to at most BIN and last four digits in the output (SKILL.md) and to treat the found data as evidence only without persistence of full numbers.
Audit Metadata