From recoup-deals
Use when ingesting a music catalog data room, royalty statements, rights files, metadata exports, or messy catalog spreadsheets for a deal review. Triggers include "catalog data room", "clean royalty statements", "normalize royalties", "music rights review", "catalog ingest", "merge catalog metadata", "prepare this catalog for valuation", or seller files containing ISRCs, ISWCs, splits, PRO statements, distributor reports, contracts, recoupment schedules, or YouTube Content ID reports.
How this skill is triggered — by the user, by Claude, or both
Slash command
/recoup-deals:recoup-deal-ingestThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Turn messy music catalog data rooms into auditable hand-off artifacts for
evals/evals.jsonreferences/canonical-schema.mdreferences/cleaning-rules.mdreferences/data-room-checklist.mdreferences/deal-workspace.mdreferences/normalization.mdscripts/_helpers.pyscripts/build-file-manifest.pyscripts/build-manual-review-queue.pyscripts/calculate-concentration.pyscripts/dataroom-hygiene-scan.pyscripts/normalize-royalty-statement.pyscripts/validate-findings-evidence.pyTurn messy music catalog data rooms into auditable hand-off artifacts for deal review and valuation. The job is not to value the catalog. The job is to make the data trustworthy enough that valuation can happen.
Start here based on what the user gives you:
If the user only wants lightweight cleanup, still preserve raw files and source columns. Never silently overwrite source data.
references/deal-workspace.md when no workspace exists.scripts/normalize-royalty-statement.py; see
references/normalization.md.scripts/build-manual-review-queue.py and report the resulting
summary_line (e.g. "29 of 88 files contributed financial data;
50 require manual review").scripts/dataroom-hygiene-scan.py and
merge any high-strength matches into findings/findings.json.scripts/calculate-concentration.py
with --assumptions to evaluate the configured threshold; merge any
auto-emitted finding into findings.json.Use these artifacts unless the user asks for a different structure:
| Artifact | Purpose |
|---|---|
workpapers/file-manifest.json | Classified manifest (parse_status, likely_provider, period, currency, rights-type hint) per source file. |
workpapers/ingest-coverage.json | Coverage summary (X of Y files contributed financial data; Z require manual review). |
workpapers/concentration-analysis.json | Top-N concentration percentages per dimension; threshold-tripped flag. |
workpapers/dataroom-hygiene.json | Filenames and content patterns suggesting seller-side concealment leaks. |
findings/manual-review-queue.md | Per-file checklist of files that did not contribute financial data. |
findings/dataroom-hygiene-findings.json | Proposed process_integrity findings to merge into findings.json. |
findings/concentration-finding.json | Proposed valuation finding when concentration threshold is tripped. |
ingest-manifest.md | Explains source files, assumptions, and scope. |
data-room-inventory.csv | One row per source file. |
canonical-catalog.csv | One row per controlled work/recording candidate. |
royalty-ledger.csv | Normalized income lines across statements. |
rights-map.csv | Ownership, splits, contracts, restrictions, support level. |
source-lineage.csv | Field-level or row-level source traceability. |
missing-files.md | Required files, unresolved conflicts, and open seller requests. |
data-quality-report.md | Profiling results and cleanup decisions. |
After producing the ledger, always run:
scripts/build-manual-review-queue.py — surfaces the X-of-Y coverage line and per-file actions.scripts/dataroom-hygiene-scan.py — surfaces concealment-language matches like DELETE_BEFORE_SHARING.txt.scripts/calculate-concentration.py --assumptions — auto-emits a finding when concentration trips the materiality threshold.Merge the proposed findings into findings/findings.json with real
evidence_ids so they survive the validate-findings-evidence.py check.
Detailed schemas are in
references/canonical-schema.md.
references/data-room-checklist.md.references/canonical-schema.md.references/cleaning-rules.md.references/deal-workspace.md.references/normalization.md.npx claudepluginhub recoupable/skillsProvides UI/UX resources: 50+ styles, color palettes, font pairings, guidelines, charts for web/mobile across React, Next.js, Vue, Svelte, Tailwind, React Native, Flutter. Aids planning, building, reviewing interfaces.
Searches MemPalace before answering questions about past work, people, projects, or prior decisions. Returns verbatim stored content instead of guessing from model memory.