notion-research-documentation
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's primary workflow involves retrieving information from external sources (Notion pages) that may contain untrusted data or adversarial instructions that could influence the agent's behavior during the synthesis phase.\n
- Ingestion points: Untrusted data enters the agent context via
Notion:notion-fetchas described inSKILL.mdandreference/advanced-search.md.\n - Boundary markers: The provided research templates (e.g.,
reference/research-summary-template.md,reference/comprehensive-report-template.md) do not implement delimiters or explicit instructions for the AI to ignore embedded commands within the source text.\n - Capability inventory: The agent has the capability to write to the workspace using
Notion:notion-create-pages, creating a risk where injected instructions could lead to the generation of malicious or misleading documentation.\n - Sanitization: There is no mention of text sanitization or filtering for the content retrieved from
Notion:notion-fetchbefore it is processed by the model.
Audit Metadata