todays-focus
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources such as task trackers and calendars, creating a potential surface for indirect prompt injection.
- Ingestion points: The agent reads from 'AGENTS.md', weekly notes, task tracker integrations, and calendar services.
- Boundary markers: A 'Read-only boundary' is explicitly defined, directing the agent to leave vault files, tasks, and calendars unchanged, and specifically keep 'index.md' and 'log.md' untouched.
- Capability inventory: Capabilities are restricted to reading and summarizing information; the skill is prohibited from file creation or tool invocation, reinforced by the 'disable-model-invocation: true' configuration.
- Sanitization: No explicit sanitization or filtering logic is provided for the content ingested from external integrations.
Audit Metadata