three.ws Autopilot
Set autopilot scopes and daily $THREE spend caps, then propose, execute, and undo agent actions.
Verdict not yet evaluated for this tool. The semantic screen takes adversarial cases first; coverage rolls out as the corpus expands (15/150 labels to graduation). The deterministic conformance probe is built but has not yet run on the public corpus, so a recorded verdict here is REVIEW or UNVERIFIED, never a clearing ALLOW. Until a verdict is recorded, an agent should treat this tool as not-yet-cleared and fall back to its own checks. Method: the eval, four-state verdict, honest limits.
Own this server? Screen its description →
THREE_WS_API_KEYREQUIRED. The agent owner's three.ws credential — a three.ws API key (sk_live_… / sk_test_…) or an OAuth access token. Sent as Authorization: Bearer on every request; every autopilot endpoint is owner-scoped and rejects unauthenticated calls. This credential can move real $THREE via execute_proposal — treat it like a password. THREE_WS_TOKEN / THREE_WS_BEARER are accepted aliases.
THREE_WS_BASEBase URL of the three.ws API that serves the autopilot endpoints. Override only when self-hosting or targeting a preview deployment.
THREE_WS_TIMEOUT_MSPer-request timeout in milliseconds. Proposal generation runs a real server-side LLM pass, so the default is generous.
AI-powered trading strategy development: backtesting, market data, and portfolio analysis
Feature flagging and A/B testing platform with AI-first experimentation workflows.
Query Meta Ads performance data — accounts, campaigns, ad sets, ads, metrics & settings.