{
  "slug": "io-github-zizka-ai-sl-zizkadb-mcp",
  "name": "ZizkaDB",
  "auditedAt": "2026-11-02T12:22:26Z",
  "version": "0.1.12",
  "ecosystem": "pypi",
  "package": "zizkadb-mcp",
  "disclaimer": "Automated smoke test, not a security guarantee.",
  "complete": true,
  "checks": {
    "versionPin": {
      "status": "pass",
      "detail": "Pinned pypi zizkadb-mcp@0.1.12",
      "version": "0.1.12"
    },
    "vulnerabilities": {
      "status": "pass",
      "detail": "No critical, high, or medium advisories were returned.",
      "critical": 0,
      "high": 0,
      "medium": 0,
      "low": 0,
      "packagesScanned": 1
    },
    "license": {
      "status": "pass",
      "detail": "MIT",
      "spdx": "MIT"
    },
    "installScripts": {
      "status": "pass",
      "detail": "No preinstall, install, postinstall, or prepare script.",
      "scripts": []
    },
    "secrets": {
      "status": "pass",
      "detail": "No credential patterns in the published artifact.",
      "findings": []
    },
    "packageSize": {
      "status": "pass",
      "detail": "11216 bytes",
      "bytes": 11216
    },
    "activity": {
      "status": "pass",
      "detail": "Active 2 days ago.",
      "ageDays": 2
    },
    "sandbox": {
      "status": "pass",
      "detail": "MCP handshake completed. 13 tools listed.",
      "starts": true,
      "handshake": true,
      "timeToReadyMs": 899,
      "tools": [
        {
          "name": "log_event",
          "description": "\nLog an event to ZizkaDB.\n\nUse this every time your agent takes an action: a tool call, a decision,\na user message, or an agent response. Logging builds the causal graph\nthat powers why(), search_memory(), and get_context().\n\nArgs:\n    agen"
        },
        {
          "name": "reset_log_chain",
          "description": "Start a fresh causal chain (new session_id, clear parent_id)."
        },
        {
          "name": "search_memory",
          "description": "\nSearch over all logged agent events.\n\nFinds past decisions, tool calls, conversations, or any event. Works on any\nserver: keyword search without embeddings, keyword + meaning (hybrid) with\nthem. Useful before a new task to see what happene"
        },
        {
          "name": "verify_chain",
          "description": "\nProve an agent's history is untouched.\n\nRechecks every event's checksum and its link in the agent's hash chain.\n\"intact\": true means no event was edited, removed or reordered (GDPR\ndeletions show up as tombstones, not breaks). Keep \"chain_"
        },
        {
          "name": "get_context",
          "description": "\nGet a formatted memory context block ready to inject into a system prompt.\n\nThis is the drop-in replacement for LLM-provided memory (Claude Projects,\nChatGPT memory). It combines recent events with semantically relevant past\nevents, fits w"
        },
        {
          "name": "why",
          "description": "\nTrace the causal chain that led to an event.\n\nGiven any event ID, walks back through parent_id links to reconstruct\nthe full decision tree: user message -> tool call -> result -> response.\nEssential for debugging why an agent did something"
        },
        {
          "name": "query_events",
          "description": "\nList recent events for an agent.\n\nArgs:\n    agent:      Agent identifier to query\n    limit:      Number of events to return (default 20)\n    event_type: Filter by event type, e.g. \"tool_call\" (optional)\n\nReturns:\n    List of events ordere"
        },
        {
          "name": "time_travel",
          "description": "\nReplay the exact state of an agent at a specific point in time.\n\nReconstructs what the agent knew and had done up to that moment.\nUseful for investigating complaints (\"what did the agent tell them at 3pm?\")\nor comparing behaviour across ve"
        },
        {
          "name": "get_baseline",
          "description": "Behavioral baseline and drift score for an agent."
        },
        {
          "name": "get_token_usage",
          "description": "Token usage and cost report for an agent over a date range."
        },
        {
          "name": "get_token_optimization",
          "description": "Deterministic token optimization suggestions from real usage events."
        },
        {
          "name": "memory_diff",
          "description": "\nSummarise what happened in a session.\n\nReturns event counts, types seen, causal depth, whether any errors\noccurred, and event types that had never appeared in prior sessions.\nCall this at the end of a session to understand what changed.\n\nA"
        },
        {
          "name": "forget",
          "description": "\nDelete all events matching a filter (GDPR right to erasure).\n\nRemoves events from both the database and the vector index.\nCommonly used to delete all data for a specific user on request.\n\nArgs:\n    filter_key:   Field in event data to matc"
        }
      ],
      "toolCount": 13,
      "resourceCount": 0,
      "promptCount": 0,
      "stderrExcerpt": "[11/02/26 09:22:26] INFO     Processing request of type            server.py:733\n                             ListToolsRequest                                   \n                    INFO     Processing request of type            server.py:733\n                             ListResourcesRequest                               \n                    INFO     Processing request of type            server.py:733\n                             ListPromptsRequest",
      "network": "disabled",
      "networkAttemptedWhileOffline": false,
      "outboundHosts": [],
      "requiresCredentials": false
    }
  },
  "tier": "gold",
  "tierReason": "Sandbox handshake completed and the published checks passed.",
  "logExcerpt": "[11/02/26 09:22:26] INFO     Processing request of type            server.py:733\n                             ListToolsRequest                                   \n                    INFO     Processing request of type            server.py:733\n                             ListResourcesRequest                               \n                    INFO     Processing request of type            server.py:733\n                             ListPromptsRequest"
}