skills/simota/agent-skills/tempo/Gen Agent Trust Hub

tempo

Pass

Audited by Gen Agent Trust Hub on Apr 30, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references external URLs for the purpose of retrieving official calendar data and documentation.
  • In references/business-calendar.md, it points to https://www.cao.go.jp/chosei/shukujitsu/syukujitsu.csv, which is the official Japanese Cabinet Office source for public holidays.
  • In references/cron-patterns.md, it mentions crontab.guru and cronhub.io, which are standard web-based tools for cron expression validation.
  • [REMOTE_CODE_EXECUTION]: The skill provides numerous code snippets in various languages (Python, TypeScript, Go, Java, Ruby) within its reference files. These are strictly for documentation and implementation guidance for the developer/agent to use during code generation. There are no patterns suggesting the skill attempts to execute remote code or download scripts for immediate execution.
  • [DATA_EXFILTRATION]: No evidence of sensitive file access (e.g., .ssh, .aws, .env) or unauthorized network transmission of data was found. The skill follows the 'store UTC, render local' discipline and emphasizes secure handling of timestamps and idempotency keys.
  • [PROMPT_INJECTION]: The instructions are technical and domain-specific. There are no attempts to override system prompts, bypass safety filters, or use adversarial 'jailbreak' patterns.
  • [COMMAND_EXECUTION]: The skill provides examples of platform-specific configuration (GitHub Actions YAML, Kubernetes CronJob manifests, etc.), but these are provided as 'ready-to-paste' snippets for the user or builder agent rather than being executed by the skill itself.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 30, 2026, 10:14 AM