qrspi-create-structure-outline
Pass
Audited by Gen Agent Trust Hub on Jun 21, 2026
Risk Level: SAFEPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [PROMPT_INJECTION]: Potential for indirect prompt injection. The skill ingests content from design and research documents to generate outputs without using boundary markers or explicit instructions to ignore embedded commands.
- Ingestion points: SKILL.md (Step 1, Step 3).
- Boundary markers: Absent; no delimiters used to isolate untrusted content.
- Capability inventory: File reading (design/research docs) and file writing (structure docs).
- Sanitization: Absent; no content filtering or validation performed on input files.
- [DATA_EXFILTRATION]: Potential for unauthorized file access. The skill is instructed to follow file references found inside the user-provided design document to read additional research data. This pattern could be exploited to trick the agent into reading sensitive local files if the path in the untrusted design document is malicious.
Audit Metadata