jfrog-mcp-management

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (low risk: 0.10). In jfrog-mcp-management runtime Step 2 (--list-available / --inspect) the agent executes npx --yes @jfrog/agent-guard ... --inspect/--mcp <MCP_NAME> and ingests JFrog catalog output (JSON/TSV) provided by the platform, not outsider-authored free text submitted into a queue/feed the agent reads.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill invokes npx @jfrog/agent-guard at runtime with --registry <REGISTRY_URL>, which defaults to https://releases.jfrog.io/artifactory/api/npm/coding-agents-npm/ — causing npx to fetch and execute remote package code that the skill relies on.