launchdarkly-flag-create
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- EXTERNAL_DOWNLOADS (LOW): The skill depends on the @launchdarkly/mcp-server package. While LaunchDarkly is a known service provider, the organization is not included in the trusted whitelist. The severity is set to LOW because this dependency is essential for the skill's primary functionality.
- PROMPT_INJECTION (LOW): The skill contains an indirect prompt injection surface (Category 8) because it instructs the agent to analyze arbitrary files within the codebase. Evidence: (1) Ingestion points: The agent reads manifest files like package.json and source code in Step 1. (2) Boundary markers: No delimiters or protective instructions are used. (3) Capability inventory: The agent can perform flag operations via MCP and has write access to local files. (4) Sanitization: No sanitization is performed on ingested code content before processing.
- COMMAND_EXECUTION (LOW): Step 5 of the workflow encourages the agent to run project-specific build or linting commands. This involves executing arbitrary commands from the local environment, which is a common but sensitive developer tool capability.
Audit Metadata