Docs | BlockSentinel

Documentation

Static integration preview for early protocol, security, and risk teams. Full hosted docs will expand as the API hardens.

BlockSentinel turns live lending-protocol stress signals into explainable exploit-risk scores, alerts, and multisig-ready guard proposals.

Getting started

Apply for a pilot and name the Ethereum lending protocol.
Share contract addresses, markets, and alert recipients.
We confirm signals, thresholds, and allowlisted guard actions.
You receive sample alerts and proposal packages.

Protocol onboarding checklist

What we need from you versus what BlockSentinel runs.

Input	Notes
RPC	Not required from the customer for a pilot if BlockSentinel runs ingest infrastructure.
Protocol contract addresses	Event emitters (e.g. Pool) plus any guard targets you want allowlisted.
Event signatures / topic0	Borrow, Repay, LiquidationCall (computed from ABI, not guessed).
Alert recipients	Slack webhook and/or email list, with severity preferences.
Threshold preferences	Risk 7d trigger, confidence floor, and dedupe window.
Optional Safe / multisig info	Safe address only if you want proposal packaging later. Not required for JSON.
Guard allowlist preferences	Which defensive actions are even allowed to be proposed.

API preview

Read-only risk API. Authentication details are issued per pilot.

GET /v1/protocols
GET /v1/risk/{proto_id}
GET /v1/risk/{proto_id}/history?hours=168

Alert payload preview

{
  "channel": "slack",
  "severity": "high",
  "proto_id": "aave_v3",
  "title": "BlockSentinel Risk Alert — Aave V3",
  "risk_7d": 8.4,
  "confidence": 0.82,
  "summary": "Oracle deviation persisted 38m; liquidation spike z=3.1.",
  "recommended_action": "Freeze WBTC reserve (allowlisted, proposal only)",
  "action_id": "BS-ETH-AAVE-2026-00019",
  "status": "PENDING"
}

Guard proposal schema preview

{
  "version": "1.0",
  "action_id": "BS-ETH-AAVE-2026-00019",
  "proto_id": "aave_v3",
  "chain": "ethereum",
  "risk_snapshot": { "risk_7d": 8.4, "risk_30d": 6.1, "confidence": 0.82 },
  "trigger_factors": [
    {
      "factor": "Oracle instability",
      "severity": "High",
      "evidence": { "oracle_dev_pct_max_1h": 2.8, "persisted_minutes": 38 }
    },
    {
      "factor": "Liquidation spike",
      "severity": "Medium",
      "evidence": { "liq_spike_z": 3.1 }
    }
  ],
  "proposed_action": {
    "type": "FREEZE_ASSET",
    "scope": "ASSET",
    "asset_symbol": "WBTC",
    "expires_minutes": 120,
    "safe_ready": false,
    "note": "JSON package for multisig review. Safe packaging is optional."
  },
  "status": "PENDING",
  "created_ts": 1770000000
}

JSON is the Phase 1 artifact. Encoded calldata and Safe drafts are optional and only for allowlisted actions.

External requirements

BlockSentinel ingest needs an Ethereum RPC, protocol emitter addresses, and event topic0 hashes. Guard proposals additionally need allowlisted target contracts and ABIs for any encoded calldata.

For a founder-led pilot, BlockSentinel can operate the RPC and ingest path. Customers still confirm addresses, alert destinations, and which defensive actions may be proposed.

Contact

Integration questions: support@blocksentinel.ai