productivity-ai-tools
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists of purely informational markdown files that provide best practices, risk assessments (sharp edges), and validation checklists for administrative tasks like email and calendar management.
- [NO_CODE]: There are no executable scripts, shell commands, or external dependencies included in the skill. All instructions are text-based grounding for the AI agent.
- [PROMPT_INJECTION]: The instructions provided in SKILL.md are standard grounding prompts that direct the agent to use the reference files as a source of truth. They do not contain any instructions to bypass safety filters, disclose system prompts, or disregard safety guidelines.
- [DATA_EXFILTRATION]: No network operations, credential usage, or sensitive file access patterns were identified. The skill does not attempt to send data to external servers.
- [INDIRECT_PROMPT_INJECTION]: While the skill mentions processing meeting notes and emails, it does not include tools or scripts for data ingestion, nor does it provide a mechanism to execute commands based on that data, effectively having no exploitable attack surface.
Audit Metadata