alirezarezvani/claude-skills / finance/skills/passive-deal-screener
passive-deal-screener
VerifiedUse when a passive investor shares a private deal (syndication, preferred equity, bridge fund, private credit) to screen as an LP. Stress-tests it, computes fee drag with fee_drag_calculator.py and compares terms with benchmark_comparator.py.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 9.4 KB |
| evals/evals.json | config | 13.2 KB |
| examples/README.md | markdown | 1.2 KB |
| examples/equity-syndication/input.md | markdown | 734 B |
| examples/equity-syndication/output.md | markdown | 19.9 KB |
| examples/hard-money-fund/input.md | markdown | 608 B |
| examples/hard-money-fund/output.md | markdown | 15.9 KB |
| examples/private-credit-fund/input.md | markdown | 676 B |
| examples/private-credit-fund/output.md | markdown | 17.0 KB |
| references/01-asset-class-norms.md | markdown | 12.8 KB |
| references/02-fee-stack-library.md | markdown | 16.9 KB |
| references/03-red-flag-library.md | markdown | 18.2 KB |
| references/04-question-bank.md | markdown | 16.5 KB |
| references/05-benchmark-returns.md | markdown | 15.4 KB |
| references/data/README.md | markdown | 3.2 KB |
| references/data/etf-comparators-snapshot.md | markdown | 7.3 KB |
| references/data/fred-10yr-snapshot.md | markdown | 2.6 KB |
| references/data/ncreif-npi-snapshot.md | markdown | 3.1 KB |
| references/data/preqin-vintage-note.md | markdown | 3.4 KB |
| scripts/benchmark_comparator.py | script | 20.0 KB |
| scripts/fee_drag_calculator.py | script | 18.5 KB |
references/data/README.md
View on GitHub# `references/data/` — versioned source snapshots
The external-source data that `05-benchmark-returns.md` synthesizes. **Every
number in `05` traces to a dated, source-cited snapshot here**, so a refresh
means re-pulling these files, not rewriting `05`'s prose — and a benchmark can't
be invented, only sourced.
## Files
| File | Source | As of | What it holds |
|---|---|---|---|
| `fred-10yr-snapshot.md` | FRED `DGS10` / `DGS2` / `DGS3MO` (St. Louis Fed), direct CSV pull | 2026-09-24 | Treasury curve — 10yr risk-free anchor + 3mo/2yr short points for duration-matching debt deals |
| `etf-comparators-snapshot.md` | Computed from Yahoo Finance dividend-adjusted daily closes | Q2 2026 quarter-end (2026-06-30) | Trailing 5yr/10yr total returns, vol, worst drop, and expense ratios for the 8 public comparators |
| `ncreif-npi-snapshot.md` | NCREIF, via IREI's report of the quarterly release | Q2 2026 | NPI total / income / appreciation returns — the institutional private-RE baseline |
| `preqin-vintage-note.md` | Preqin, via CAIS's public citation | 2001–2017 study window (latest public; re-confirmed 2026-09-27) | Categorical note on private-fund vintage-year net-IRR dispersion |
## Sources — four, one per kind of data
| Data | Source | How |
|---|---|---|
| Treasury curve | **FRED** | Direct CSV: `https://fred.stlouisfed.org/graph/fredgraph.csv?id=DGS10,DGS2,DGS3MO` |
| ETF returns, vol, worst drop | **Yahoo Finance** | Public chart endpoint, dividend-adjusted daily closes; all figures computed from the one price history |
| NCREIF NPI | **IREI** | Its report of each NCREIF quarterly release |
| Private-fund dispersion | **CAIS** | Public citation of Preqin data (Preqin's own tables are subscription-gated) |
Two figures sit outside the quarterly cycle and are confirmed **once a year**:
ETF expense ratios (on each fund company's page) and the NPI full-year and
property-type returns (from the annual recaps). A documented gap is a correct
output; a fabricated number is a defect.
## Refresh procedure
1. **Treasury curve** — download the FRED CSV above; take the latest daily close
for each of the three series.
2. **ETF figures** — for each of the eight tickers, pull the full daily history
from Yahoo and compute, as of the latest quarter-end: trailing 5yr and 10yr
annualized total return, 10yr annualized vol, and full-history worst drop
(method in `etf-comparators-snapshot.md` → Provenance).
3. **NCREIF NPI** — take the latest quarter's total / income / appreciation and
the trailing four-quarter total from IREI's report.
4. **Preqin** — re-confirm the CAIS citation; if a Preqin subscription is
available, replace the categorical note with current quartile boundaries.
5. Update each file's `As of` / `Captured` stamp, then update `05`'s table and
its own `LAST_UPDATED`.
6. Update the hardcoded copies in `scripts/benchmark_comparator.py`
(`COMPARATORS`, `TREASURY`, `LAST_UPDATED`) and its `_self_check` anchors,
then run `python scripts/benchmark_comparator.py --self-check`. Its drift
guard fails if any constant disagrees with `05`.
Annual refresh is the minimum cadence; refresh sooner if a rate regime shifts
materially (the 10yr anchor moves the whole illiquidity-premium calculation).
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| finishing-a-development-branchobra/superpowersUse when implementation is complete, all tests pass, and you need to decide how to integrate the work | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| systematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixes | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| test-driven-developmentobra/superpowersUse when implementing any feature or bugfix, before writing implementation code | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| skill-creatoranthropics/skillsCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |