Revenue Recovery

---

name: revenue-recovery description: Core revenue recovery orchestration for Xzenia. Use when onboarding a client, running end-to-end revenue leakage analysis, coordinating data ingestion, contract parsing, causal analysis, recovery planning, executive reporting, or checking client recovery status.

---

Revenue Recovery

Coordinate these local skills:

contract-parser

financial-telemetry

causal-engine

recovery-actions

executive-briefing

Workflow

1. Create client workspace in workspace/data/xzenia/clients// 2. Load or write client_config.json 3. Ingest source data 4. Parse contracts 5. Run causal analysis 6. Generate recovery plan 7. Generate executive briefing 8. Persist outputs and summarize findings

Orchestration scripts

scripts/onboard-client.sh [client_name] [arr] [billing_system] [billing_model]

scripts/run-client-cycle.sh [csv_file]

These provide the current subsystem entrypoints for onboarding and full-cycle execution.

Governing texts

This skill operates under the following bound artifacts, in precedence order: 1. projects/xzenia/docs/co-architecture-directive.md 2. projects/xzenia/docs/operational-reliability-charter.md 3. projects/xzenia/docs/the-third-scroll.md 4. projects/xzenia/docs/the-scroll.md 5. projects/xzenia/state/trust-expansion-rubric.md 6. projects/xzenia/state/self-audit-ledger.md

Rules

Never fabricate findings.

Report data gaps explicitly.

Do not execute external billing/customer actions without approval.

Prefer durable artifacts over persuasive language.

Treat external validation as optional; treat evidence as mandatory.

Proceed automatically on safe local next steps; pause only at genuine risk boundaries.

Founder vector outranks local optimization.

Never fabricate findings.

Report data gaps explicitly.

Do not execute external billing/customer actions without approval.

Prefer durable artifacts over persuasive language.

Treat external validation as optional; treat evidence as mandatory.

Proceed automatically on safe local next steps; pause only at genuine risk boundaries.

Founder vector outranks local optimization.