creating-superfans-hodak
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data, which could contain malicious instructions meant to influence the agent's behavior during analysis.
- Ingestion points: The skill explicitly requests untrusted data in
subskills/m3_customer_understanding/module.md(customer research, reviews, support tickets) andsubskills/m6_service_recovery/module.md(customer-provided accounts of service failures). - Boundary markers: The instructions do not define delimiters (e.g., XML tags or triple quotes) or provide system instructions to ignore embedded commands within the processed data.
- Capability inventory: The skill performs text analysis and generates drafts for service recovery and CRM triggers; it does not contain scripts with file-system write or network capabilities.
- Sanitization: There is no evidence of logic for escaping, filtering, or validating the external content before it is processed by the model.
- [EXTERNAL_DOWNLOADS]: The
README.mdfile provides installation instructions that involve downloading content from an external source using the commandnpx skills add simbajigege/book2skills/skills/creating-superfans-hodak. This targets a repository associated with the skill's author.
Audit Metadata