Allows verification of local model access via Ollama as part of vendor exit readiness drills.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Trust Gate MCP mint receipt for approving high-value purchase"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Post-quantum, tamper-evident receipts for consequential agent actions, as an MCP server.
Seven tools, one shared signing primitive: the open-source OpenAgentOntology mint_receipt.
What actually gets signed depends on what you install -- OAO detects its backend at import time, so this is worth stating plainly rather than advertising the best case:
Install | Legs | Notes |
| Ed25519 + ML-DSA-65 (FIPS 204) | Default. Pure Python ( |
| Ed25519 + ML-DSA-65 + SLH-DSA (FIPS 205) | Adds the hash-based diversity leg via |
PQ-required verify (the default) demands at least one verified post-quantum leg, so the
dual-leg default is a real post-quantum posture, not a downgrade -- but only the [slh] install
gives you the hash-based third leg.
Tool | What it does |
| Mints a post-quantum receipt for a CRM record change. Works with any CRM (open-core Relaticle, hosted CRMs via their own MCP, custom). Old/new values are SHA-256 hashes. |
| Ranks a CALLER-PROVIDED list of MCP tools by worst-regret if they act. Read-only. Cannot auto-discover other servers -- MCP protocol does not allow that. |
| Post-quantum receipt for any consequential agent action. |
| Verify a receipt from the certificate alone -- offline, no DB. Defaults to PQ-required mode. |
| Two-phase decision gate. PREVIEW returns risk assessment + preview_id without acting. COMMIT verifies inputs match and mints a tamper-evident receipt with execution permit. |
| Egress classification. Scans data for sensitivity markers and classifies as PUBLIC / INTERNAL / CONFIDENTIAL / RESTRICTED. Blocks RESTRICTED. Returns classification + retention info + receipt. |
| Vendor exit readiness drill. Checks local signing key, local model access (Ollama), and local data export. Returns step-by-step results + receipt. Informational, no side effects. |
H1 key persistence + bootstrap with FAIL-CLOSED kid-drift check
H2 per-IP token-bucket rate limit (DoS-hardened: FIFO eviction + body cap)
H3 PQ-required verify (defeats signature-stripping downgrade attacks)
H4 128-bit kid on every minted receipt (offline same-notary check)
Optional bearer-auth toggle + narrowed CORS via TRUST_GATE_BEARER_TOKEN + TRUST_GATE_ALLOWED_ORIGINS
33/33 tests including adversarial PQ-strip + IP-rotation attack simulations
See PUBLISH.md for the full hardening status table.
Related MCP server: DCL Evaluator
pip install trust-gate-mcp
trust-gate-mcpAdd [slh] for the hash-based third leg. From a checkout, pip install -e ".[dev]" then
python -m trust_gate_mcp.
docker build -t trust-gate-mcp .
docker run -p 8081:8081 -v trust-gate-data:/data/oao trust-gate-mcpThe volume mount on /data/oao is required for production -- without it the signing key rotates per restart and breaks long-running verification chains. The persistent key_metadata.json holds the notary's kid; the bootstrap step refuses to start if it drifts.
Apache-2.0. Built on the open-source OpenAgentOntology primitive.
This server cannot be installed
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
Bitcoin-anchored, tamper-evident audit log for AI agents — record, disclose and verify actions.
Pre-action attestation perimeter for AI agents — 8 primitives, signed C18 receipt per call.
Permission boundary receipts for ChatGPT agents.
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/CWNApps/trust-gate-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server