hermesplant.com

AI agent action safety: one-cent deterministic preflight for every consequential agent command. Returns risk, matched rule ids, and an exact machine-readable escalation call: full DestructGuard evidence for medium risk or the complete Action Safety workflow for high and critical risk.

Deterministic portfolio risk analysis and quality scoring. Computes riskScore, riskLevel, volatility, sharpe, maxDrawdown, concentrationHHI, diversification, effectiveHoldings plus per-rule findings with fixes and sector/position breakdown. Pure math only from caller holdings and returns. No external data.

Deterministic risk scorer for autonomous-agent commands. Scores a shell, SQL, or git action for destructive, deployment, and data-mutation risk before execution and returns a risk level, the matched policy rules, and an approve-or-block recommendation. Pure rules, evidence per match, no fabrication.

Analyze an MCP server manifest for security risk before install — destructive actions, over-broad scopes, sensitive-data egress, weak auth, prompt-injection surface, and permission diff. Returns per-tool, per-factor findings with severity, why, and a concrete fix (not just a score).

Submit an autonomous-agent command for paid risk triage and human-approval routing. Returns a deterministic risk decision, a request id, whether a human is required, and a recommended action — pause-and-escalate on critical risk. Evidence-backed, nothing fabricated.

Deterministic cash flow analysis and return analytics in one call: NPV, IRR (periodic), XIRR (dated), MOIC / DPI / TVPI, payback period, plus an optional DCF valuation (enterprise value, equity value, implied share price) with Gordon-growth or exit-multiple terminal value. Pure math from caller-provided cashflows — no market-data feed, no fabrication. Every metric is accompanied by evidence-backed findings and a signed response receipt.

Wallet risk scoring and pre-transaction AML, sanctions, counterparty-risk, and crypto compliance screening for autonomous finance agents. Scores one wallet from caller-provided sanctions hits, exposure tags, counterparty labels, transfers, and fund-source context; returns a 0-100 risk score, evidence per signal, and a concrete action. Deterministic preflight only: no chain data is fetched and no screening result is fabricated.