DCL Trust Oracle
The DCL Trust Oracle is a cryptographically secured AI audit layer that evaluates LLM outputs for policy compliance, safety, quality, and data leakage, storing only metadata in a tamper-evident chain. Services are metered via x402 micropayments (USDC on Base).
Capabilities
Pre-Action Evaluation:
dcl_evaluate_fast($0.01): Quick policy check.dcl_evaluate_strict($0.05): Deep analysis with higher confidence.dcl_evaluate_jailbreak($0.02): Detects prompt injection and role-hijacking.dcl_evaluate_safety($0.01): Baseline screening for harmful patterns.dcl_evaluate_quality($0.03): Content quality and drift check.dcl_evaluate_batch($0.10): Bulk evaluation of up to 20 items.
Post-Action Detection:
dcl_evaluate_secrets($0.02): Scans for credentials (API keys, JWTs, private keys, DB URLs, etc.).dcl_evaluate_pii($0.02): Detects personal data (emails, phone numbers, credit cards, IPs, passport numbers, etc.) with Luhn validation.
Session Management:
dcl_pipeline_start($0.05): Start a long-running audit session for continuous drift tracking (modes: NORMAL, WARNING, ESCALATION, BLOCK).
Audit & Forensics:
dcl_audit_decode($0.10): Retrieve basic audit record by transaction hash.dcl_audit_decode_deep($0.50): Deep forensic audit with chain integrity and drift context.
Integration: Available as an MCP server and an HTTP REST API for flexible AI agent integration.
Policy Customization: Use built-in policies or define custom policies inline.
Free Utilities: Non-metered endpoints for service health, policy listing, and audit chain status/export.
Tamper-Evident Storage: Raw content is never stored; only SHA-256 hashes and metadata are recorded in an SQLite-backed chain, ensuring verifiable integrity.
DCL Trust Oracle — x402 MCP Server
Deterministic AI audit layer with cryptographic micropayments via x402 protocol.
What It Does
DCL Trust Oracle provides deterministic policy evaluation for LLM outputs with a tamper-evident audit chain. The system stores only cryptographic hashes and decision metadata — never raw content — enabling verifiable, post-action forensic analysis across distributed AI agents.
Available two ways, both metered per call via x402:
REST API (
webhook_server.py) — direct HTTP integration, x402-gated withfastapi-x402.MCP Server (
mcp_server.py) — native Model Context Protocol integration for AI agents, hosted on Smithery, x402-gated withpaymcp.
Both servers share the same evaluation logic and tamper-evident chain (dcl_core.py), and are priced identically.
Related MCP server: Mcp-Omega-Brain
Quick Start
REST API
pip install -r requirements.txt
python webhook_server.pyServer runs on http://localhost:8080
MCP Server
pip install -r requirements.txt
python mcp_server.pyServer runs on http://localhost:8081 (streamable-http transport)
Tools & Endpoints
Pre-Action Evaluation
REST Endpoint | MCP Tool | Price | Description |
|
| $0.01 | Fast policy check for low-risk outputs. Returns tamper-evident |
|
| $0.05 | Deep analysis for high-stakes outputs with higher confidence thresholds. |
|
| $0.02 | Instruction adherence check — detects prompt injection patterns and role-hijacking attempts. |
|
| $0.01 | Baseline screening for known harmful text patterns. Optimized for high throughput. |
|
| $0.03 | Content quality & drift check — evaluates format adherence and contextual drift. |
|
| $0.10 | Bulk processing — up to 20 items per transaction. Cost-effective for multi-turn history. |
Session Management
REST Endpoint | MCP Tool | Price | Description |
|
| $0.05 | Initializes a long-running audit session for continuous drift tracking. Returns |
Post-Action Forensics
REST Endpoint | MCP Tool | Price | Description |
|
| $0.10 | Basic post-action audit — returns verdict, confidence, agent_id, reason by |
|
| $0.50 | Deep forensic audit — includes drift context, tamper-evidence indices, environmental metadata. |
Utility (free, REST only)
Endpoint | Description |
| Service status and chain length |
| List of built-in policy names |
| Chain integrity, drift mode, drift score |
| Full chain export with integrity verification |
Example Response
{
"verdict": "COMMIT",
"confidence": 0.95,
"reason": "All policy checks passed",
"tx_hash": "0x7a8f3b2c...",
"chain_index": 42,
"input_hash": "0x9d4e1f...",
"policy_version": "1.0.0",
"timestamp": 1721635200.123,
"pipeline_id": "abc123",
"drift_mode": "NORMAL",
"drift_score": 0.15
}x402 Integration
All paid endpoints and tools require payment via the x402 protocol before returning a result — no free tier, no bypass. Both the REST API and MCP server settle to the same wallet.
Networks: Base, Avalanche, IoTeX
Asset: USDC
Facilitator (REST):
https://x402.org/facilitator(viafastapi-x402)Facilitator (MCP):
paymcpinMode.AUTO— automatic on-chain payment for x402-aware MCP clients, guided payment link (ELICITATION/RESUBMIT) for clients without a wallet. No path skips payment.
Metadata-Only Architecture
DCL Trust Oracle is designed around a hash-based audit trail:
Raw content never stored — only SHA-256 hashes and decision metadata
Tamper-evident cryptographic chain (SQLite-backed, WAL mode)
Chain survives server restarts — audit records persist indefinitely
Verifiable integrity via
GET /chain/statusFull export via
GET /chain/exportShared chain across REST and MCP — a
tx_hashreturned by one interface can be audited through the other
Each audit record contains: tx_hash, prev_hash, verdict, input_hash, policy_hash, agent_id, reason, confidence, task_type, timestamp, and drift_context.
Built-in Policies
Policy | Min Confidence | Purpose |
| 0.70 | General-purpose evaluation |
| 0.80 | Detects prompt injection, role-hijacking, DAN-style attacks |
| 0.75 | Baseline harmful-pattern screening |
| 0.85 | Quality assurance, format adherence, drift detection |
Custom policies can be passed inline as YAML via the policy field.
Links
Live API: https://fronesislabs.com/docs
MCP Server: https://mcp.fronesislabs.com
MCP Manifest: https://fronesislabs.com/.well-known/agent.json
Smithery.ai: https://smithery.ai/servers/fronesislabs/dcl-trust-oracle
Glama.ai: https://glama.ai/mcp/servers/Fronesis-Labs/dcl-webhook
Contact
Maintenance
Related MCP Servers
- AlicenseAqualityCmaintenanceAI agent provenance, trust, and auditability layer. VERITAS multi-gate scoring, Cortex approval gates, S.E.A.L. hash-chain audit ledger, and semantic RAG with cryptographic provenance tracking for every decision an agent makes.Last updated275MIT
- Alicense-qualityBmaintenanceCryptographic proof of every AI decision. An immutable, verifiable audit trail MCP server.Last updated1MIT
- Alicense-qualityCmaintenanceMCP server providing immutable audit logging, policy enforcement, and compliance reporting for AI agent workflows, enabling regulatory compliance and chain integrity verification.Last updatedMIT
Related MCP Connectors
Runtime AI governance: decision gates, human approval, hash-chained audit, compliance mapping.
Bitcoin-anchored, tamper-evident audit log for AI agents — record, disclose and verify actions.
Reputation oracle for AI agents on Base: SAFE/CAUTION/BLOCK + 0-100 score before you pay. x402+MCP
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/Fronesis-Labs/dcl-webhook'
If you have feedback or need assistance with the MCP directory API, please join our Discord server