risk-register-first-pass
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill integrates with an external Model Context Protocol (MCP) server at
https://account.ismscopilot.com/v1/account/mcpto perform compliance framework lookups. This is a third-party service provided by ISMS Copilot, which is used to assist the agent in identifying specific risk criteria. - [INDIRECT_PROMPT_INJECTION]: The skill has a broad data ingestion surface that includes reading external inputs such as incident logs, penetration test findings, and customer security questionnaires.
- Ingestion points: Analyzes local files containing scope statements, asset inventories, architecture overviews, vendor lists, incident logs, audit findings, and policies (
SKILL.md). - Boundary markers: No specific delimiters or boundary markers are defined for the ingested data content.
- Capability inventory: The skill can read local project files, write new CSV and Markdown files to the local filesystem, and perform network requests to an external API.
- Sanitization: The skill contains explicit rules to prevent data exfiltration, instructing the agent to never include file contents, secrets, or personal data in external requests, and requiring the agent to show questions to the user before sending them to the external service.
Audit Metadata