A secure\* runtime for autonomous AI agents, where security policy is derived from a human-readable constitution. _\*When someone writes "secure," you should immediately be skeptical. What do we mean by secure?_
@provosMarkdown formatting conventions for email summary documents — heading depth, list style, line length, emoji policy, and a…
@provosReference for Gmail's search query syntax — operators like is:sent, newer_than:, from:, has:attachment, label:, and how they…
@provosReference vocabulary for designing instrumented harnesses that drive vulnerability discovery — design classes (trigger-driven vs…
@provosCanonical shape of the .workflow/emails/emails.json file passed between the fetch and summarize states — required fields (sender,…
@provosReference vocabulary for memory-safety vulnerabilities in native C/C++ code — bug-class taxonomy, common arithmetic patterns that…
@provosTone and length conventions for email summaries — voice, verb tense, what to include vs omit, and target sentence count. Read…
@provosUse this agent when you need to implement a design, feature, or refactor code with an emphasis on clean architecture,…
@provosUse this agent when designing new features, planning component interfaces, defining type hierarchies, or architecting…
@provos© 2026 Flowy
Built for Claude Code · Not affiliated with Anthropic
