morning-briefing
Pass
Audited by Gen Agent Trust Hub on Jun 30, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill operates as a localized personal assistant. While it ingests data from local workspace files such as raw logs and schedules, its primary function is to summarize and prioritize this information for the user without making external network calls.
- [COMMAND_EXECUTION]: The skill utilizes file system operations through an MCP (Model Context Protocol) tool to read project metadata and write briefing caches to the
ai-workspace/50-artifacts/briefings/directory. These actions are transparently documented as part of the skill's workflow and are restricted to the local workspace environment. - [DATA_EXFILTRATION]: The skill analyzes potentially sensitive user data, including daily schedules and active task lists. Analysis of the instructions confirms that this data is used exclusively to populate the local briefing provided to the user; no patterns for remote transmission or unauthorized data access were identified.
- [SAFE]: The skill incorporates self-correction measures, such as the 'Completeness Gate' and 'Safety Rules,' which require the agent to disclose missing sources and separate documented facts from inferences, reducing the risk of hallucinations or misrepresentation of work status.
Audit Metadata