content-weekly-plan
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONTIME_DELAYED_CONDITIONAL
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data from Notion to generate content recommendations and automated summaries, creating a surface for injection.
- Ingestion points: Reads from Notion databases (
Strategy,Idea Pool,Posts DB) to obtain pillars, rhythm, and content ideas. - Boundary markers: The instructions do not define delimiters or warnings to ignore potentially malicious instructions embedded within the Notion content.
- Capability inventory: The skill is capable of writing new pages to Notion and sending content to a Telegram channel via a configured digest.
- Sanitization: There is no specified logic for sanitizing or escaping the data retrieved from Notion before it is used to construct recommendations or messages.
- [TIME_DELAYED_CONDITIONAL]: The skill includes hardcoded date-specific logic that modifies how content is filtered.
- Evidence:
Personal-test для П'ятниці (жорсткий гейт, 17.09.2026). - Context: The instructions implement a mandatory content quality gate specifically linked to the date 17.09.2026. While this appears to be a business logic versioning rule, hardcoded date-based triggers can be used to conditionally alter agent behavior.
Audit Metadata