xbert-config-audit

{
  "name": "xbert-config-audit",
  "displayName": "XBert Config Audit",
  "description": "Is XBert configured consistently across your book? Spot inconsistencies, gaps, and drift in one pass.",
  "author": {
    "name": "XBert Intelligence",
    "email": "[email protected]"
  },
  "homepage": "https://intelligence.xbert.io/plugins/xbert-config-audit",
  "license": "MIT",
  "keywords": [
    "practice-config",
    "xbert-config",
    "audit",
    "quarterly",
    "consistency",
    "drift"
  ],
  "x-xbert": {
    "displayName": "XBert Config Audit",
    "tagline": "Is XBert configured consistently across your book? Spot inconsistencies, gaps, and drift in one pass.",
    "audience": [
      "principals",
      "practice-managers"
    ],
    "categories": [
      "practice-config",
      "xbert-config",
      "audit",
      "quarterly"
    ],
    "prerequisites": [
      "XBert account with multiple client tenants in a Connect",
      "Workflow and XBert configurations active on at least some clients",
      "Practice owner or manager-level access to the Connect",
      "XBert Working Paper, the companion render plugin that writes the Word document — install it from this same catalogue (it needs Python 3.10+)"
    ],
    "screenshots": [],
    "demoVideo": null,
    "includes": {
      "skills": 1,
      "commands": 1,
      "mcpTools": 0
    },
    "longDescription": "Your XBert rule set should look the same across clients in the same segment. In practice it almost never does — different staff onboarded different clients, defaults drifted as the rule library grew, and 'always-on' rules ended up on most clients but not all. Config Audit is the quarterly pass that surfaces that drift before it costs you.\n\nThe plugin reads the per-client XBert configuration across your Connect, fingerprints each rule's enablement pattern, and reports three things: rules enabled inconsistently within a segment, rules that should be on everywhere but are missing on N clients, and rules that have been switched off (drift). Then it pairs each finding with a bulk-normalisation move so the cleanup is concrete, not abstract. This is distinct from XBert Workflow Review — that plugin looks at templates and schedules; this one looks at XBert rule configuration itself.",
    "useCases": [
      {
        "title": "Quarterly config sweep",
        "description": "Once a quarter, run the audit to catch drift before it becomes systemic — clients onboarded last quarter often miss rules added recently."
      },
      {
        "title": "Post-onboarding consistency check",
        "description": "After a batch of new clients comes on, confirm they all have the rule set the practice considers baseline."
      },
      {
        "title": "Pre-audit / pre-review preparation",
        "description": "Before an internal QA pass or external review, make sure the practice can demonstrate consistent XBert configuration across the book."
      },
      {
        "title": "Standardisation campaign",
        "description": "When the practice decides to standardise on a new ruleset, use the audit to drive the rollout — see who needs what enabled and what disabled."
      }
    ],
    "benefits": [
      {
        "icon": "GitBranch",
        "title": "Inconsistency, gap and drift in one pass",
        "description": "Three distinct fingerprints — inconsistencies inside a segment, gaps where always-on rules are missing, and drift where rules have been switched off."
      },
      {
        "icon": "ShieldCheck",
        "title": "Bulk-normalisation moves named",
        "description": "Every finding is paired with the specific bulk update that resolves it, so cleanup is a list of decisions, not a research project."
      },
      {
        "icon": "FileText",
        "title": "Quarterly-ready Word output",
        "description": "Word document with executive summary, segment-by-segment breakdown and the normalisation backlog — ready to action with the team."
      }
    ],
    "workflow": [
      "Type /config-audit in Claude Code or Claude Desktop",
      "Confirm the Connect tenant and the client segmentation rule (default = by industry tag, fallback = whole book)",
      "Plugin pulls the Connect account review, configuration assignments and rule list via XBert MCP",
      "Per-segment fingerprints are computed, inconsistencies / gaps / drift are classified",
      "Build the working-paper payload, render via xbert-working-paper:render-docx, and report the saved path"
    ]
  }
}