lov-skill-pricing
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes external skill source code, README files, and capability briefs to generate pricing cards. This processes untrusted data which could contain malicious instructions designed to bias the pricing evaluation or influence the agent's behavior during analysis.\n
- Ingestion points: The agent is instructed to read files from the target skill's root directory, including SKILL.md, README.md, and manifests (SKILL.md Step 1).\n
- Boundary markers: The instructions do not specify the use of delimiters or 'ignore embedded instructions' prompts when processing the ingested skill data.\n
- Capability inventory: The skill's workflow is focused on data analysis and report generation; it does not explicitly invoke subprocesses, perform network exfiltration, or execute arbitrary code beyond its stated analytical purpose.\n
- Sanitization: The skill implements a 'Private context' separation step (Step 2) to ensure that sensitive elements like source prompts or internal paths are treated as judgment context rather than being directly included in the final output card.
Audit Metadata