openjutsu
One-click character swap for videos: an open-source, local Higgsfield Genjutsu alternative built on ComfyUI and MiniMax H3, with FLUX.3 Klein relighting.
Dated, hash-chained calls on whether failed startup categories are ripe again: the constraint that killed each one, the metric that moved since, and a sourced graveyard of funded companies.
MediaPython · JavaScriptGitHub @theguysaccountOnlineOwner unconfirmed
Have you tried it? One tap is enough. No sign-in needed.
No updates detected yet. GitHub releases will appear here.
Owners of public repositories under a personal GitHub account can sign in to verify ownership and manage the introduction themselves. No administrator approval is needed.
Sign in, generate a verification file, and commit it to this repository’s default branch within 30 minutes. Files on other branches or forks do not count. Then check the file below to manage the project without administrator approval. Keep the file to retain your access.
If needed, you will go to sign-in. After signing in, return here and press your chosen verification button again.
Run /nomorevibe verify in your project folder to verify ownership and manage the introduction yourself.
Verify in one minuteAsk about other verification options
Ripeness board: dated, hash-chained calls on whether dead startup categories are viable again, with sourced graveyards and live public metrics.
Taken from the repository README
Loading the full README.
Read the full README ↗Live: https://theguysaccount.github.io/idea-locator-prophet-fund/ · Ledger: calls, verdicts and resolutions · Data (CC BY 4.0): board.json, graveyard.csv, slips.csv · Screens: docs/screens
A public, dated, hash-chained record of calls on whether failed startup categories are viable again. Each category names:
The rule behind it: a category is worth re-attempting when funded companies already burned real money proving the demand and the specific constraint that killed them has since crossed a threshold. Webvan and Instacart are the canonical pair. The method page has the backtests and every test that failed.
Also on the site:
| Job | Workflow | When | Fails safe because |
|---|---|---|---|
| Refresh metrics (FRED, World Bank, OWID), the Federal Register calendar and ClinicalTrials.gov; resolve calls whose rule can now be decided; commit; redeploy | refresh.yml | Mondays 06:17 UTC | each source fails independently and keeps its last good data; tests run before any commit |
| Test, guard the ledger, build, render share images, deploy to Pages, ping IndexNow | deploy.yml | on content or code changes, after a refresh, or by hand | tests, the build or the ledger guard failing stops the job; Pages keeps the last good deploy |
| Live health check, source link check, tests on main | selftest.yml | Thursdays 05:41 UTC | read-only checks |
| Keep-alive: re-enable schedules, heartbeat commit after 40 quiet days | keepalive.yml | called by every scheduled workflow | runs even when the calling job fails |
| Tests on every push and pull request (Python 3.9 and 3.12) | ci.yml | every push | n/a |
Details, alerts and the keys a human can turn later: OPERATIONS.md.
il/ Python 3.9+, standard library only
data.py load and validate data/ (the build refuses invalid data)
ledger.py append-only hash chain: canonical JSON, verify, prefix check
register.py write new calls and verdict changes onto the ledger
refresh.py fetch public data, build the calendar, run automatic resolvers
resolve.py resolve a manual call with evidence (used by the scheduled review)
site.py static site generator (pages, embeds, share-card HTML, sitemap, data exports)
charts.py server-rendered SVG charts
data/
categories/ one JSON file per category: verdict, constraint, metric, graveyard, evidence, sources
calls.json calls to register (once registered, the ledger is the source of truth)
slips.json the slip corpus
backlog.json candidate categories for the scheduled review to verify
live/ fetched data (metrics, calendar, trials, run log), committed by refresh.yml
ledger/chain.jsonl the ledger
site/ CSS, browser scripts (share, ledger verifier, re-look), static page bodies, IndexNow key
scripts/ share images (og.py), ledger guard, IndexNow, health and link checks, screenshots
tests/ unittest suites + Node tests for the browser scripts
python3 -m il.refresh # fetch public data and resolve due calls (no keys needed)
python3 -m il.site # build into dist/
python3 scripts/og.py # optional: render share images with local Chrome
python3 -m http.server 4731 --directory dist
python3 -m unittest discover -s tests -t .
node --test tests/js.test.mjs # after a build; checks the browser verifier and re-look matcher
Deploys happen on push to main (deploy.yml). To deploy by hand: gh workflow run deploy.yml.
data/categories/<slug>.json, then run python3 -m il.register. A verdict change is appended to the ledger with today's date. Then run the tests.data/calls.json and run python3 -m il.register. Registered calls can never change; to revise a view, add a new call.python3 -m il.resolve <id> yes|no|void --evidence "..." --url https://....The deploy job refuses to publish a ledger that does not extend the one already live (scripts/guard_ledger.py).
IL_BASE_URL, IL_CUSTOM_DOMAIN, IL_GOATCOUNTER (cookieless analytics), IL_RELOOK_REPORT_URL (the money slot). See .env.example and OPERATIONS.md.
Code MIT; content and data CC BY 4.0. See LICENSE. Research, not investment advice.