infrahub-sot

Pass

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill documentation references an external repository at github.com/opsmill/infrahub-mcp for the MCP server implementation.
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface due to its interaction with untrusted infrastructure data.
  • Ingestion points: Data is ingested from Infrahub through get_nodes, get_related_nodes, and query_graphql.
  • Boundary markers: There are no explicit markers or instructions provided to the agent to treat retrieved infrastructure data as untrusted or to ignore embedded instructions.
  • Capability inventory: The skill allows for state-changing operations including GraphQL mutations via query_graphql and the creation of infrastructure branches via branch_create.
  • Sanitization: No data sanitization or schema validation processes are specified for content retrieved from the Infrahub instance.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 6, 2026, 12:31 AM