qa-investigation

Pass

Audited by Gen Agent Trust Hub on Sep 1, 2026

Risk Level: SAFEDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill provides a methodological framework for debugging test failures and does not include any executable scripts or remote dependencies.
  • [DATA_EXFILTRATION]: The skill incorporates security best practices by explicitly instructing the agent to redact sensitive information (tokens, cookies, credentials, PII) before writing findings to the filesystem (SKILL.md, references/flow.md).
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data which constitutes an attack surface. 1. Ingestion points: The agent reads external test logs, stack traces, and application traces as described in SKILL.md and references/flow.md. 2. Boundary markers: The instructions do not define specific delimiters for ingested content in the investigation files. 3. Capability inventory: The skill performs file writing to qa_investigation_plan.md, qa_investigation_findings.md, and qa_investigation_progress.md. 4. Sanitization: The skill mandates the redaction of sensitive data and summarizing multimodal content into text, which mitigates some risks of raw data injection.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 1, 2026, 03:51 AM
Security Audit — agent-trust-hub — qa-investigation