flow-nexus-swarm
Warn
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the user to install the
flow-nexuspackage from the NPM registry vianpm install -g flow-nexus@latest. This is a third-party dependency from a source not identified as a trusted vendor.\n- [REMOTE_CODE_EXECUTION]: The skill provides instructions to execute remote code from bothflow-nexusandclaude-flowpackages usingnpxfor account registration, login, and managing the MCP server environment.\n- [INDIRECT_PROMPT_INJECTION]: The skill's workflow processing logic is vulnerable to indirect prompt injection.\n - Ingestion points: External data enters the agent context via the
input_dataparameter in themcp__flow-nexus__workflow_executetool call within SKILL.md.\n - Boundary markers: There are no explicit delimiters or system instructions provided to ensure the agent distinguishes between workflow data and embedded instructions.\n
- Capability inventory: The platform includes powerful capabilities such as spawning new agents (
mcp__flow-nexus__agent_spawn), orchestrating tasks (mcp__flow-nexus__task_orchestrate), and executing multi-step deployment workflows.\n - Sanitization: The instructions do not define any sanitization, validation, or filtering mechanisms for the external data ingested into the swarm.
Audit Metadata