Agent safety & governance MCP servers

6 runtime tools: 2 HemmaBo tools, 2 host onboarding tools, and 2 VRP verification tools. Not an OTA.

Builds S1000D data modules from structured drafts, verifies them, applies only unambiguous fixes.

Audits shipped apps for how easily they can be reverse engineered

Read-only IBKR MCP: account, positions, brief, regime, stress, reporting, Edge, and risk.

Evidence-readiness MCP server: validate, audit, and score briefs, memos, and evidence packs.

Construct, inspect, visualize, and simulate evidence-backed ODE models with BioMASS

AI hardware synthesis: CircuitBlocks DSL, live LCSC parts, industrial DFX audit, stdio MCP server.

Read-only connectors; local writes. Evidence and assessments for agents. Only people approve.

Codebase understanding stored with its evidence, marked stale when that evidence changes.

Neutral tie-breaks between AI agents and sealed review of an agent's output, with signed receipts.

Make AI apps answer your way, check with you before acting, and get approvals on your phone.

Human in the loop for AI agents: ask ID verified people to check a decision before acting.

Read S1000D data and publication modules as structure, with applicability and verification status.

Validate S1000D data modules: XSD, BREX chain and reference resolution as one verification status.

Banco Central do Brasil (BCB): SGS series, Focus expectations, PTAX, stats + provenance. 17 tools.

Local cognitive identity engine with traits, behavioral evidence, and revision history.

Accountable autonomy MCP server: gating, tamper-evident audit for every action

Accountable autonomy proxy: gating, tamper-evident audit, time anchor for every action

MCP proxy: holds risky tool calls for passkey approval; signed receipts you can check offline.

Free desktop CAE: import CAD/STL, solve FEM structural/thermal and CFD, verify the answer.

LLM cost optimization gateway, repo token-waste auditor & model arbitrage MCP server.

Needs the PrivacyFence app from privacyfence.eu: human approval and audit for AI access to your data

Epoch time estimation MCP server for PERT, COCOMO, Monte Carlo, calibration, cost, and risk.

Local-first YNAB MCP for reconciliation, receipt splitting, safe cleanup, and spending analysis

Model, measure and verify parametric parts and assemblies in a running SolidWorks (Windows).

Free x402 offer/receipt checker: verifies signatures and whether the signer fits the host.

Verify AgentBook registration and operator attestations for agent wallets on World Chain.

Free WCAG 2.2/ADA/508 accessibility MCP: scan, AI fixes, verify, vision audit, localhost tunnel

Founders' AI COO: every AI cofounder's Routines on time — heartbeat, failures, approvals, reports.

Deadline-aware governance for AI agents: PII scan always, safety check if it fits the deadline

Safety-first OPNsense firewall MCP: read-only by default, planned and auto-reverting writes.

Stdio MCP: 17 verifiable AIMarket oracles, 35 pay-per-call tools for AI agents.

The record layer for the machine age: counter-signed receipts and agent action records.

Policy enforcement plane for MCP: every tool call ends in a verdict. Self-hosted, MIT.

Audit your own smart-home devices: destinations, encryption, open ports, A-F grade.

Scans MCP server tool definitions for excessive agency, injection sinks, hardcoded secrets

Baselines an MCP server's tools and detects tool-definition changes after approval

Scans MCP server auth code for OAuth 2.1 anti-patterns: token passthrough, missing audience

Vet link sources before you submit, verify listings, and audit your backlinks for link spam.

Notify people through Cue: preferences, attention budgets, quiet hours and approval apply.

Adopt, read, verify and record an Open Entity identity (OEIS-1). No auth, no CAPTCHA.

Drafts a network allow list for agent tools and runs them in a network-blocked sandbox.

50 starter MCP tools for AI compliance and trust: identity, escrow, SLA, EU AI Act.

Manage MagicSword endpoints, alerts, policies, and private intelligence through MCP.

From first signal to satellite evidence: live world-event intelligence and collection planning.

Witness layer for agents: signed receipts binding a fingerprint to a witnessed time. $0.001.

Gateway in front of MCP tools: critical arguments are checked against trusted sources and a policy.

Runs coding-agent workers on a codebase question; every claim in the answer carries evidence.

Approval gate and audit trail for AI agent actions: policy check first, proof after. Any MCP client.

Read-only MCP for FINOS AI governance risks, controls and crosswalks. Independent project.