defold-project-build

Pass

Audited by Gen Agent Trust Hub on Sep 23, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill automates project builds by communicating with the Defold editor's local HTTP API. All network activity is restricted to the local loopback address (127.0.0.1).
  • [INDIRECT_PROMPT_INJECTION]: The skill processes output from the Defold editor, creating a surface for potential instruction injection via project source code or build errors.
  • Ingestion points: Responses from the /command/build and /console endpoints (SKILL.md).
  • Boundary markers: None. The skill does not instruct the agent to treat editor output as untrusted data or use delimiters.
  • Capability inventory: Network requests to localhost via curl and Invoke-RestMethod, and local file system access to .internal/editor.port.
  • Sanitization: None. The content retrieved from the editor is processed directly.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 23, 2026, 02:42 PM