LaunchDarkly Automation
Pass
Audited by Gen Agent Trust Hub on Mar 29, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as it retrieves data from LaunchDarkly environments and projects which could theoretically influence the agent's behavior.
- Ingestion points: Project, environment, and code reference data retrieved via
LAUNCH_DARKLY_LIST_PROJECTS,LAUNCH_DARKLY_GET_ENVIRONMENTS, andLAUNCH_DARKLY_LIST_CODE_REFERENCE_REPOSITORIES(SKILL.md). - Boundary markers: Absent. The instructions do not define delimiters for external data or advise the agent to disregard instructions embedded within the fetched content.
- Capability inventory: The skill includes tools to create and delete trigger workflows (
LAUNCH_DARKLY_CREATE_TRIGGER_WORKFLOW,LAUNCH_DARKLY_DELETE_TRIGGER_WORKFLOW), allowing for state changes in the external service (SKILL.md). - Sanitization: Absent. There is no indication of input validation or output sanitization for the data retrieved from LaunchDarkly before it is used in subsequent operations.
- [EXTERNAL_DOWNLOADS]: The skill directs users to connect to an external MCP server at
https://rube.app/mcp, which is the official endpoint for the vendor's integration service.
Audit Metadata