2 projects
tool-call-guard
Deny-by-default policy gate for AI agent tool calls: allowlists, argument validation, rate caps, human-approval hooks, dry-run mode, and an audit trail. Framework-agnostic, zero dependencies.
llm-context-budget
Fit chat history into any model's context window: token estimation, sliding-window packing with a running summary, and overflow detection. Zero dependencies, no AI inside.
