safety governance
2 articles · 4 co-occurring · 0 contradictions · 5 briefs
LLM outputs from MCP interactions must be handled carefully to prevent security problems like cross-site scripting (XSS) or other web application attacks if the output is shown to users" — Article dir
2026-W15 10
LLM outputs from MCP interactions must be handled carefully to prevent security problems like cross-site scripting (XSS) or other web application attacks if the output is shown to users" — Article dir
@tokenbender: the trajectory is set - the models would learn to write a minimal scaffold gi... extends
[INFERRED] "humans would teach how to whitelist/blacklist based on policies" — Article describes human-defined policy boundaries (whitelist/blacklist) as a governance mechanism. Extends safety-governa
query this concept
$ db.articles("safety-governance")
$ db.cooccurrence("safety-governance")
$ db.contradictions("safety-governance")