service-itsm-teams-itservice-configure
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it retrieves and processes data from external objects like 'User' and 'Community'. Maliciously crafted data in fields such as user names or site names could potentially influence the agent's logic during the configuration process.
- Ingestion points: Data is ingested from the Salesforce org via SQL queries for users (
SELECT Id, Name, Username, Email FROM User) and API calls for communities (/connect/communities) as described inSKILL.md(Step 3 and Step 4). - Boundary markers: There are no explicit delimiters or instructions to the agent to ignore or isolate potential command strings embedded within the results of these queries.
- Capability inventory: The skill uses the
mcp__headless-360__dispatchtool to perform state-changing operations includingPATCHandPOSTrequests to Salesforce API endpoints. - Sanitization: The instructions do not provide mechanisms for sanitizing or validating the content of the data retrieved from the org before the agent processes or presents it.
Audit Metadata