# Source ledger — YC identity gate study

All dates are observation or retrieval dates, not publication dates. The table preserves historical research-input hashes except S5, which identifies the current published primary packet. Some historical inputs are not redistributed; current downloadable artifacts are linked below. Local absolute paths are omitted. No private data or credentials were used.

| ID | Source | Date | Hash / receipt | Use | Limits |
| --- | --- | --- | --- | --- | --- |
| S1 | `portfolio-status.json` | generated 2026-09-21 | SHA-256 `f0ec40d722d88b391dde27dda933d5be9de3340f1844dae88c71afc999fb2a4c` | Exact selection of qualified idea `accelerator-trends-w2-r4-027`; original score vector and evaluation receipt | Qualification is eligibility, not completed research; `publication_linkage` was empty. |
| S2 | `yc-company-rosters.json` | retrieved 2026-09-18 | SHA-256 `e18917eb02e8338ba579c7cf2d974d68a97e41781a0d5f0f1f0a523bbf2fb406` | All 1,586 current public YC directory rows across nine cohorts; row ID, cohort, listed website, public one-liner | Current public index coverage, not original historical intake coverage; no demographic or employment inference. |
| S3 | `homepage-link-inventory.json` | observed 2026-09-19 | SHA-256 `7926b9993db471ca2457c52ba02c6c303acd578cdf5115def38bbbd59f2d3336` | Saved 1,586-row guarded homepage inventory; final URLs, statuses, HTML-known flag, robots/error state, and bounded initial text excerpts where available | Raw HTTP/HTML signal only; bounded excerpts may miss JavaScript-only content; no full response bodies, reference-page fetch, rendered-browser claim, product execution, ownership, or adoption claim. |
| S4 | `analysis.json` | analyzed 2026-09-19 | SHA-256 `b317c2d1e5437f7bff46fbd8b5cbd8e4d5d43027fd742d236a85a76f48b23dbc` | Existing independently checked denominators and observation-range metadata; used as a cross-check, not as a new network run | Its headline topic is buyer-reference links; this study reuses the saved row/final-URL observation for a separate identity gate. |
| S5 | [Current primary verification](../yc-identity-gate-primary-verification-2026-09-21.json) | fetched and policy-rechecked 2026-09-21 | SHA-256 `f2cb08d3a5c0969da47bd624e796cfa2014f180ec16ac1ce3f34bab6a0b126b0` | Bounded re-fetch of only the 39 saved host-change final homepages; 39/39 status 200, no errors; title/headings/meta description and body SHA-256 | Credential-free public GET plus robots check; response body discarded; title/token signal is escalation evidence, not legal identity. One public personal statement was omitted from the packet and not used. |
| S6 | `content/articles.json` | current registry at review | SHA-256 `618756048b93c106b4ffc32f13ad202e5172dde9bfa405df3e59a6be6fe5fd52` | Overlap audit of 45 non-company editorial records | Existing company profiles were not treated as duplicate research articles; near-neighbor editorial records are listed in `selection.json`. |
| S7 | First-party YC cohort pages: [Summer 2024](https://www.ycombinator.com/companies?batch=Summer%202024), [Fall 2024](https://www.ycombinator.com/companies?batch=Fall%202024), [Winter 2025](https://www.ycombinator.com/companies?batch=Winter%202025), [Spring 2025](https://www.ycombinator.com/companies?batch=Spring%202025), [Summer 2025](https://www.ycombinator.com/companies?batch=Summer%202025), [Fall 2025](https://www.ycombinator.com/companies?batch=Fall%202025), [Winter 2026](https://www.ycombinator.com/companies?batch=Winter%202026), [Spring 2026](https://www.ycombinator.com/companies?batch=Spring%202026), [Summer 2026](https://www.ycombinator.com/companies?batch=Summer%202026) | source snapshot 2026-09-18 | URLs retained in S2 rows | Public source identity and cohort filters | Directory can change after the snapshot. |
| S8 | Primary examples: [Hey Revia YC row](https://www.ycombinator.com/companies/hey-revia) → [OpenLoop Health](https://openloophealth.com/); [Midship YC row](https://www.ycombinator.com/companies/midship) → [Optro](https://optro.ai/); [Storia AI YC row](https://www.ycombinator.com/companies/storia-ai) → [current public page](https://juliaturc.com/); [Maive YC row](https://www.ycombinator.com/companies/maive) → [Ladder](https://ladderai.com/); [Permitify YC row](https://www.ycombinator.com/companies/permitify) → [Pangea](https://pangeanotes.com/) | verified 2026-09-21 | Per-response hashes in S5 | Concrete manual-review examples | These links show public page identity signals only; no acquisition/rebrand/legal conclusion is made. |
| S9 | Same-brand signal examples: [Wardstone row](https://www.ycombinator.com/companies/wardstone) → [wardstone.space](https://wardstone.space/); [Kontigo row](https://www.ycombinator.com/companies/kontigo) → [kontigo.com](https://www.kontigo.com/); [YouShift row](https://www.ycombinator.com/companies/youshift) → [youshift.com](https://youshift.com/) | verified 2026-09-21 | Per-response hashes in S5 | Counterexamples showing why cross-domain is a gate, not a verdict | Same-brand title signal does not prove ownership, current status, or adoption. |

## Current downloadable files

- [Offline recount code](method.py) makes no network requests.
- [Recount output](../yc-identity-gate-recount-2026-09-18.json) contains all cohort totals, 34 cross-domain row records, and aggregate counts for the five same-registrable-domain changes.
- [Primary verification](../yc-identity-gate-primary-verification-2026-09-21.json) contains all 39 host-change records, including those five subdomain changes. It retains only public metadata, response sizes/hashes, policy evidence and bounded identity signals, not page bodies.
- [Derived summary](../yc-identity-gate-2026-09-18.json) contains the study findings.
- [Public compact inventory](../yc-homepage-links-2026-09-19.json) provides the 1,586-row input for the public recount.

Use the portable command in [the methods README](README.md). An optional archived roster must match the compact inventory's names, cohorts and IDs. The bounded network verifier accepts caller-supplied input/output paths and stores metadata only; a future run is a new observation.