hl-spike
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest a user-provided topic and generate technical experiments, creating a surface for indirect prompt injection where a malicious topic could attempt to guide the agent toward unauthorized actions.
- Ingestion points: The
[topic]input inSKILL.md. - Boundary markers: Absent; there are no specific delimiters or instructions to ignore embedded commands within the topic.
- Capability inventory: The skill facilitates command execution and observation through the "Run safe, available experiments" instruction.
- Sanitization: Absent; the skill does not define validation or sanitization logic for the input topic.
Audit Metadata