when-analyzing-user-intent-use-intent-analyzer

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt explicitly captures process.env and filesystem/context into the agent's stored "userInput" and memory (e.g., userInput.context: { environment: process.env }) which would expose environment variables and any embedded secrets to the LLM and allow them to be included verbatim in outputs.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This skill deliberately captures sensitive local context (process.env, cwd, recent edits, filesystem) and routes those artifacts into memory, disk, and external agent/hook invocations (npx claude-flow agent-spawn/hooks and embedding APIs), creating clear avenues for data exfiltration and credential theft to remote services.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill repeatedly invokes the remote npm package via runtime npx commands (e.g., "npx claude-flow@alpha" in multiple hooks/agent-spawn/skill-run lines), which will fetch and execute remote code at runtime and is relied on to spawn agents and control prompts, so this is a required external runtime dependency that can execute code.