phase-plan

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly requires pasting the "Original User Request (verbatim)" into the generated plan, which would force the model to reproduce any secrets or API keys present in the triggering prompt.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The skill's required workflow (SKILL.md) mandates running /skill-oracle "to discover relevant skills across the full local catalog (~244 skills) and ClawHub registry" and to "Use the output to: Keep only implementable skills in the plan" — meaning the agent will fetch and interpret content from a public third‑party registry (ClawHub) and use it to drive skill selection and subsequent actions, exposing it to untrusted third‑party instructions.