Plugins
OpenClaw plugin exposing agent-optimizer's audit/plan/apply/rollback/scan verbs as agent tools. Mutating tools are approval-gated.
Did that actually work? Did my message actually go out? Is this safe to send? Why did my request fail? Should I stop and ask? Am I going in circles? What did I lose in compaction? 22 answers for OpenClaw agents from a local ledger of every request and tool outcome. Verify skills before install; see where installed skills send data; catch silent failures (tool said ok, wire said 403); confirm message delivery against the channel API; keep secrets from leaving. No cloud, no telemetry, observe-only by default, MIT.
61 specialist workflows and a Chief of Staff with local recovery and evidence tools.
HashiCorp Vault integration for OpenClaw - secure secret management with SecretRef support, conversational skill, and CLI helper
Pauses risky OpenClaw actions, like rm -rf, force-push, destructive SQL and prod deploys, until you approve them in Telegram or on the OKed web dashboard. Everything else just runs. Every decision gets a signed receipt.
