manufacturing-ai-app-builder
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill uses the
npx skills@latest addcommand to fetch and install delegate skills from the author's GitHub organization (open-edge-platform). These downloads are targeted at the vendor's own repositories to provide additional functionality for vision and sensor processing tasks. - [COMMAND_EXECUTION]: The orchestrator utilizes shell commands including
lsto verify the presence of local skill files andsedto extract and inspect metadata from skill headers. It also relies ongitandghfor repository operations and interacting with the GitHub platform, as defined in its allowed-tools configuration. - [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied business objectives to determine routing and technical parameters. This introduces a surface for indirect prompt injection if external data is used to influence the agent's behavior. However, this is mitigated by a mandatory manual verification step in the procedure (Step 4), which requires the agent to present a plan and wait for human approval before any build or installation commands are executed.
Audit Metadata