continuous-learning-v2

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

• Third-party content exposure detected (high risk: 0.80). The skill's CLI explicitly supports importing instincts from arbitrary HTTP(S) URLs (scripts/instinct-cli.py cmd_import) and SKILL.md documents the /instinct-import workflow, so external, untrusted web content can be fetched, parsed into instincts, and then applied/promoted to influence agent behavior.