creating-agent-skills

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

• Third-party content exposure detected (high risk: 0.80). The skill's references (references/api-security.md) explicitly instruct the agent to call external services (e.g., via curl or ~/.claude/scripts/secure-api.sh for "facebook", "yourservice", etc.) and to read/act on those API responses as part of the workflow, which exposes it to untrusted third‑party content that could carry indirect prompt-injection instructions.