kano-model
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data by instructing the agent to use retrieval tools to gather competitor information and reviews, which could potentially contain malicious instructions.
- Ingestion points: External evidence gathered via retrieval tools and user-provided context in SKILL.md.
- Boundary markers: No explicit instructions are provided to the subagents to ignore or isolate embedded instructions within the retrieved external data.
- Capability inventory: The skill has the capability to write files (generating the markdown report) and orchestrate multiple subagents as described in SKILL.md.
- Sanitization: The skill lacks explicit sanitization or filtering logic for processing external content before it is included in the final report.
- [NO_CODE]: The skill consists entirely of markdown-based instructions and reference documents. There are no included scripts, binaries, or complex automation files to audit for code-level vulnerabilities.
Audit Metadata