cli-ux-review
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill requires the agent to ingest and evaluate untrusted external data in the form of third-party CLI source code and runtime outputs, creating a potential surface for indirect prompt injection if the audited code contains malicious text instructions.
- Ingestion points: Target application source code and command execution outputs specified under the 'How to run' section in
SKILL.md. - Boundary markers: Absent. The instructions do not prescribe specific boundary delimiters or safety warnings telling the agent to ignore instructions embedded within the analyzed source files.
- Capability inventory: There are no custom scripts, subprocess calls, network commands, or file system writes bundled within this skill package.
- Sanitization: Absent. No escaping or pre-validation of the source text is outlined before it enters the agent's context window.
- [NO_CODE]: The skill bundle contains no executable code, binaries, or automated scripts. It is composed entirely of markdown documentation (
SKILL.md), metadata (DERIVATION.json), and legal terms (LICENSE), meaning it cannot perform active malicious operations independently.
Audit Metadata