community-launch-runner
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists purely of Markdown-based instructions and reference materials without any executable code, scripts, or automated tool configurations.
- [INDIRECT_PROMPT_INJECTION]: The skill identifies ingestion points for untrusted data (scraped text, reviews, comments, and emails) and includes specific boundary instructions to ignore embedded commands. The capability inventory includes connectors to various marketing and RevOps APIs, but the instructions mandate treating all external content as untrusted evidence, providing adequate sanitization and instruction-override guidance for the agent context.
- [CREDENTIALS_UNSAFE]: The skill follows security best practices by explicitly stating that API keys should not be kept in skills or source control, and no hardcoded credentials or secrets were found in the provided reference materials.
- [COMMAND_EXECUTION]: There are no shell commands, subprocess calls, or dynamic context injection patterns (e.g., the !command syntax) present in the instructions or reference files.
Audit Metadata