mcpField
H
Hive Log

Hive Log

Remote
hiveagentiq.comCloud & Infra0Updated 26w ago

About Hive Log

Hive Log is an MCP logging service for autonomous agents, with structured log ingestion and access to each agent’s own logs. Ingestion is charged per line, with selectable retention periods.

Key features

  1. 01NDJSON batch ingestion
  2. 02Tail and search logs by DID
  3. 031-day, 7-day, or 30-day retention
  4. 04Daily usage and charge summary

Use cases

  1. 01Record agent tool-call events
  2. 02Investigate logs by severity or tag
  3. 03Review daily ingestion and charges

MCP support

MCP tools provide log ingestion, retrieval, retention controls, and daily summaries. Log reads are limited to the caller’s own DID.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://hive-mcp-log.onrender.com/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

Tools

Details

From the project README

hive-mcp-log

srotzin/hive-mcp-log MCP server

Inbound structured-log ingestion shim — Hive Civilization

NDJSON log ingestion at $0.0001/line via x402. Retention tiers (1d free, 7d $0.005/MB, 30d $0.02/MB). Tail + search endpoints. Pure protocol — inbound only, no outbound calls.

Council provenance: Tier A inbound metering shim. Companion to hive-mcp-auction and hive-mcp-barter.


What this is

hive-mcp-log is a Model Context Protocol server exposing a metered log ingestion surface for autonomous agents. Agents POST structured logs (NDJSON) and pay $0.0001/line in USDC on Base L2. They read their own logs back via tail/search at Tier 0 (free). Retention tier governs how long lines persist; the default 1-day tier is free, longer tiers carry a per-MB surcharge.

  • Protocol: MCP 2024-11-05 over Streamable-HTTP / JSON-RPC 2.0
  • Transport: POST /mcp
  • Discovery: GET /.well-known/mcp.json
  • Health: GET /health
  • Settlement: USDC on Base L2 — real rails, no mock, no simulated
  • Brand gold: Pantone 1245 C / #C08D23
  • Inbound only: the service never reaches out, never DMs, never pulls

MCP tools

ToolTierDescription
log_ingestmeteredIngest NDJSON batch (max 4 MB / 10k lines). $0.0001/line + retention surcharge.
log_tail0 (free)Return last N log lines for the calling DID.
log_search0 (free)Search own-DID logs by time/severity/tag/free-text.
log_retention_get0 (free)Return current tier + bytes stored.
log_retention_set0 (free)Change retention tier (1d/7d/30d).
log_today0 (free)Today aggregate — ingests, lines, bytes, charge_usd.

REST endpoints

MethodPathPurpose
POST/v1/log/ingestIngest NDJSON batch. Returns 402 envelope first, accepts tx_hash to settle.
GET/v1/log/tailLast N lines for did.
GET/v1/log/searchSearch by time/severity/tag/q. Cursor pagination.
GET/v1/log/retentionCurrent retention tier + monthly cost projection.
PATCH/v1/log/retentionChange retention tier for a DID.
GET/v1/log/todayToday aggregate (Tier 0, free).
POST/mcpMCP JSON-RPC 2.0 surface.
GET/healthService health.

Pricing

Ingest:    $0.0001 per log line
Retention:
  1d   (default)   free
  7d               $0.005 / MB
  30d              $0.02  / MB
Tail / search (own-DID):  Tier 0 — free

NDJSON ingest example

curl -X POST https://hive-mcp-log.onrender.com/v1/log/ingest \
  -H 'content-type: application/json' \
  -H 'x-hive-did: did:hive:agent-foo' \
  -d '{
    "did": "did:hive:agent-foo",
    "retention_class": "7d",
    "lines": [
      { "ts": "2026-04-27T20:00:00Z", "severity": "info",  "tag": "tool", "msg": "tool_call ok" },
      { "ts": "2026-04-27T20:00:01Z", "severity": "error", "tag": "llm",  "msg": "rate_limited" }
    ]
  }'

The first call returns a 402 with the payment envelope (Base L2 USDC). Settle, then resend with x-hive-tx: <hash> for synchronous verification.

Tail / search example

curl 'https://hive-mcp-log.onrender.com/v1/log/tail?did=did:hive:agent-foo&n=20'
curl 'https://hive-mcp-log.onrender.com/v1/log/search?did=did:hive:agent-foo&severity=error&q=rate'

Retention

curl -X PATCH https://hive-mcp-log.onrender.com/v1/log/retention \
  -H 'content-type: application/json' \
  -d '{ "did": "did:hive:agent-foo", "tier": "7d" }'

A periodic sweep (default hourly) prunes lines older than each DID's retention window.

Environment

VarDefaultNotes
PORT3000
ENABLEtrueFlip to false to disable ingestion (returns 503 with log_ingestion_disabled).
WALLET_ADDRESS0x15184bf50b3d3f52b60434f8942b7d52f2eb436eW1 MONROE on Base L2.
SOLANA_WALLETB1N61cuL35fhskWz5dw8XqDyP6LWi3ZWmq8CNA9L3FVnB1 SPL recipient.
USDC_BASEcanonicalUSDC contract on Base.
BASE_RPChttps://mainnet.base.org
PUBLIC_BASE_URLhttps://hive-mcp-log.onrender.com
MAX_LINES_PER_INGEST10000
MAX_BYTES_PER_INGEST41943044 MB.
RETENTION_SWEEP_MS36000001 hour.

Security & scope

  • Cross-DID reads are not exposed in v1 (own-DID only via tail/search).
  • No PII detection in v1; client-side discipline assumed. Schema check is a v1.1 candidate.
  • ZK cardinality attestation is a separate shim (hive-mcp-zk-attestation); not bundled.

License

MIT — Steve Rotzin / Hive Civilization

Hive Gamification

This MCP server is part of the Hive Civilization gamification surface (10-mechanic capability taxonomy).

Surface tags: gamification.spec.v1 · gamification.surface.public · gamification.signal.read-only · gamification.settlement.real-rails

Real rails on Base L2 (USDC 0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913). Read-only signal layer. Brand gold #C08D23.

Hive Civilization Directory

Part of the Hive Civilization — agent-native financial infrastructure.

Brand: #C08D23

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.