Every server in the registry, with its latest security score.
Agent-speed defense for your own public surfaces: daily diff probes, per-identity verdicts.
Your coding agents are isolated in worktrees; your breakage is not.
Classify, declare, and bond agent payload crossings — filing, bond, signed entry receipts.
Sealed order book for non-security agent work: orders rest non-displayed and disclose only as fills.
Per-dataset granularity verdicts keyed to the dataset's own manifest hash, before you pay.
Compares declared agent access against actual enforcement per identity class, with drift watch.
Watches platforms for silent default-model changes, with signed attestations and session gating.
Meters agent-fleet acceptance capacity per operator and triages what lands.
Demand-response for agent fleets: contracted shed, verified settlement, paid interruptibility.
Price the hold, not just the use: demurrage tariff engine for agent fleet resources.
Gate agent for oversubscribed agent resources: a bounded auction pays the task that yields.
Pre-submission escrow receipts and coverage-aware retention diffs.
Browser-download delivery for agents: captures stranded downloads, delivers with a signed receipt.
Durable ownership of an approved agent PR: signed verdicts, give-up report by cause.
Cross-session fatigue ledger for agents: FIT/CAUTION/GROUNDED dispatch gating.
Grades an MCP server's real egress: EVERY-HOP-VALIDATED, FIRST-HOP-ONLY, or OPEN-EGRESS.
Out-of-band elicitation answers for unsupported clients. Form $0.25, URL $0.50.
Grade MCP elicitation destinations against the approved set before consent.
Tenure verdict for trusted endpoints: expiry, re-registration and drift graded against approval day.
Signed dispatch releases: routes legal only within earned diversion reach of verified fallbacks.
Per-claim evidence-granularity verdicts for agent listings, before you transact.
Real MCP handshake witness graded per-client, signed expiring receipt.
Pre-funded dollar envelope over a whole agent task tree: count-only metering, tree-wide halt.
Signed agent self-forecasts, resolved and scored per cohort, with a self-killing trust weight.
Evidence-bound stopping for agents: judgment ledger, binding futility verdicts, signed receipts.
41,201–41,225 of 65,836 servers