press-kit
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [SAFE]: The skill is designed for creative writing and public relations tasks. The instructions follow industry-standard best practices (e.g., inverted pyramid structure, 5-W lede) and provide templates for various announcement types like funding rounds and product launches.
- [COMMAND_EXECUTION]: The repository includes a helper script,
scripts/verify.sh, which is intended to be used by the agent or user to validate the formatting of a drafted press release. The script performs static, network-free analysis using standard Unix utilities (grep,sed,wc,tr) to check for required markers like headers, datelines, and end markers. The script handles file inputs safely with proper quoting and does not involve any dynamic code execution or network operations. - [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided product briefs and company information. While this represents a data ingestion surface, the skill's output is intended for human-readable markdown or text documents (press releases). Furthermore, the skill has no network exfiltration capabilities or privileged access, mitigating the impact of any potential injection in the source data.
Audit Metadata