Startup resources: 92 external resources, checked and credited - #6
Merged
Conversation
…dited A new sidebar section, last in the rail because it is other people's material rather than the method. Six pages plus a landing, built from two curated collections merged and de-duplicated. The source list held 186 rows and 179 unique links. Ninety-two are here. What happened to the rest is on the landing page, because an index that quietly drops half its contents is not one anyone should trust: - 68 links pointed at playbooks.cyclotronroad.org (45) and growth.500.co (23). Both fail DNS outright. I checked twice through different network paths after the first check returned an ambiguous status that turned out to be the local proxy refusing CONNECT rather than the sites being gone -- the second check was ENOTFOUND on both, which is the real answer. - 50 rows were one Google Slides deck, one row per slide. Counting it as 50 resources would have inflated the collection by a third. It is one entry. - 7 returned a hard 404 or 410. Five are Y Combinator library pages that moved. One is Third Derivative's own deal roadmapping template, now 410 Gone -- the only resource in the collection D3 authored rather than collected. Attribution follows the publisher, not the collector: a Y Combinator essay is credited to Y Combinator. Where a resource sits on a platform rather than a publication -- YouTube, Google Docs, Medium -- no publisher is named at all, because the author cannot be recovered from the link and a guess would credit the wrong person. The collection itself is credited to Third Derivative, New Energy Nexus and RMI, who assembled it. Every classification was verified rather than trusted. Each agent returned the first 40 characters of the title it had read, and those echoes were checked against the source file: 92 of 92 matched. This was not precautionary -- an earlier run of the same work was discarded because the input had been transcribed by hand and some records were wrong, which produced notes attached to the wrong resources. The echo check is what makes that failure visible instead of shippable.
|
Caution Review failedThe pull request is closed. ℹ️ Recent review info⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Plus Run ID: 📒 Files selected for processing (9)
📝 WalkthroughWalkthroughAdds an annotated startup-resources index and six category pages, registers them in the generated site map and sidebar under “Startup resources,” and updates README navigation and page-count references. ChangesStartup resources
Estimated code review effort: 2 (Simple) | ~10 minutes Possibly related PRs
✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
This was referenced Jul 29, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Adds a Startup resources section, last in the sidebar because it is other people's material rather than the method. Six pages plus a landing.
What is in it
Grouped by what the collection actually contains rather than by the stack's ten sections — a reading list and a knowledge base are different shapes.
What was dropped, and why
The source list held 186 rows and 179 unique links. Ninety-two are here.
playbooks.cyclotronroad.org(45) andgrowth.500.co(23). BothENOTFOUND. Worth noting how this was established: the first link check returned an ambiguous status for these, which I initially read as "dead". It was actually the local proxy refusingCONNECT. I re-checked through a different network path before asserting anything, and both really are gone. Cyclotron Road's playbooks have a live successor at Stanford Climate Ventures, but the paths do not map, so recovery is a title-by-title job.Attribution
Credit follows the publisher, not the collector. A Y Combinator essay is credited to Y Combinator. Where a resource sits on a platform rather than a publication — YouTube, Google Docs, Medium — no publisher is named, because the author cannot be recovered from the link and a guess would credit the wrong person. The collection itself is credited to Third Derivative, New Energy Nexus and RMI.
How the classification was verified
Each classifying agent returned the first 40 characters of the title it had actually read, and those echoes were checked against the source file. 92 of 92 matched.
That check is not decoration. An earlier run of this work was discarded because I had transcribed the input by hand into the workflow arguments and got some records wrong — which produced notes attached to the wrong resources, and would have shipped a page crediting Y Combinator for a Yale grants listing. The rewrite has agents read the file from disk so no data passes through transcription, and the echo check makes that class of failure visible rather than shippable.
Not included
The second source's site is behind Cloudflare and returns 403 to everything; the Internet Archive holds only 5 of its 42 topic pages. Those 5 pages contributed 38 resources, which are in. The rest is not retrievable and is not claimed.
node web/build.mjsclean: 68 pages, 76 search entries. README counts updated (the build fails if they drift).Generated by Claude Code
Summary by CodeRabbit
New Features
Documentation