figma-multi-env-setup
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructions involve using the
gcloudandghcommand-line interfaces to manage secrets. These tools are explicitly permitted in the skill's configuration and are used for their intended purpose of secure credential storage. - [INDIRECT_PROMPT_INJECTION]: The skill sets up a logic flow where the application's behavior is determined by external environment variables (e.g.,
NODE_ENV). While this represents a surface for indirect prompt injection where untrusted environment data could influence the agent's configuration tasks, it is a standard implementation for multi-environment configuration. - Ingestion points: Environment variables parsed in
src/config/figma.tsand defined in.envfiles (e.g.,SKILL.md). - Boundary markers: Absent; values from environment variables are trusted by the configuration logic without explicit delimiters.
- Capability inventory:
Read,Write,Edit,Bash(gcloud:*), andBash(vault:*)as defined inSKILL.mdfrontmatter. - Sanitization: No explicit sanitization or validation of environment variable values is provided in the example code snippets.
Audit Metadata