orchestrate-multi-target-sdks
Fail
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTION
Full Analysis
- Indirect Prompt Injection (HIGH): The skill facilitates a workflow where external content (OpenAPI specs) is ingested from 'registry.speakeasyapi.dev' and processed to generate code with automated side effects (GitHub Pull Requests via 'mode: pr'). This creates a vulnerability where malicious instructions or schemas in the spec could compromise the generated SDK or the CI/CD environment.
- External Downloads (MEDIUM): The skill references an external GitHub Action 'speakeasy-api/sdk-generation-action'. Per security policy, this source is not listed as a trusted organization, posing a risk of supply chain attack if the action is compromised.
- Command Execution (LOW): The skill documentation suggests the use of the 'speakeasy' CLI tool for local execution, which involves running external binaries with repository-level permissions.
Recommendations
- AI detected serious security threats
Audit Metadata