exact-online-uk
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill enables the ingestion of untrusted data from the Exact Online UK Accounting API, which represents a surface for indirect prompt injection if the agent processes data fields containing malicious instructions.\n
- Ingestion points: Data is retrieved through methods such as
apideck.accounting.invoices.listas documented inSKILL.md.\n - Boundary markers: The instructions do not specify any boundary markers or delimiters to isolate retrieved external data from the agent's core instructions.\n
- Capability inventory: The skill allows for reading and writing financial data (including invoices, payments, and journal entries) through the Apideck unified API.\n
- Sanitization: The skill provides no instructions or logic for sanitizing or validating external content before it is interpolated into the agent context.
Audit Metadata