Conversation
`ghostget pdf <url>` passed every link to Wordcell, which fetches anonymously, so papers behind a university sign-in failed even when Chrome was signed in. With --cookie-source, --browser-profile, --cookie-profile, --auth or --cookies-file, Ghostget now downloads the PDF over pinned HTTPS with the same cookie reader and per-URL filter that `read` uses, checks the PDF signature, explains a sign-in page in plain language, and hands Wordcell an owner-only temporary copy. Without these options the command is unchanged. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Pack src/pdf-auth.ts and remeasure the reviewed package budget (619 entries); the no-sign-in pdf path now decides in usage.ts and never imports pdf-auth. - Refuse private IP literals and local names before reading that host's cookies; SECURITY.md states the enforced order. - Configure the keychain notice for every run, so agents and non-TTY runs get the permission-notice line before a browser store is read. - Treat a cookie file with no cookies for a hop as an empty set, so a DOI redirect reaches the publisher. - Keep the paper's requested and final URLs as the note's source by handing Wordcell a remote source for the local copy. - Drop --mode from the signed-in options; Wordcell rejects it as before. - Plain messages for unknown accounts and cookie-file failures, sizes in MB, and exact-text tests. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Hold back SameSite=Strict cookies on redirect hops that leave the link's site. - Count every cookie read against --timeout-ms. - Stream the PDF into an owner-only file instead of buffering it twice. - Report a missing sign-in (wrong browser or profile) separately from missing access. - Resolve a browser data folder given to --browser-profile to its Default profile, and refuse folders with no sign-in. - Tell users to switch an http:// link to https://. - Accept --trust-profile-egress and --mode copied from read next to a sign-in option. - Restore the ghostgetHelpRequest doc comment. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
0thernet
enabled auto-merge (squash)
September 29, 2026 00:26
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This branch was successfully deployed
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
1 participant
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.
Summary
ghostget pdf <url>passed every link to Wordcell, which downloads anonymously, so papers behind a library or university sign-in failed even when Chrome was signed in.With
--cookie-source,--browser-profile,--cookie-profile,--auth, or--cookies-file, Ghostget now downloads the PDF itself:pinnedHttpsFetch), at most five redirects; credential-bearing, plain-HTTP, and private targets fail before any cookie is read.readuses, looked up separately for each hop, so a site never receives another host's cookies.--max-pdf-bytesand--timeout-ms; the body must start with%PDF-.no-accesserror that suggests opening the link in the browser first.pdfimport with the remaining options, then removed, including on failure.Without these options the command is byte-for-byte the previous delegation. The help topic, golden fixture, Agent Skill, and
SECURITY.mddescribe the new path. No browser profile is launched, so profile egress consent does not apply.Validation
bun run typecheck,bun run check:effect,bun run check:cost-surfaces: passbun test src/pdf-auth.test.ts src/cli.test.ts src/usage.test.ts scripts/ci-pr-gate.test.ts: 72 pass, 0 fail, "Ran 72 tests across 4 files"--cookie-source chrome(no cookie values printed): arXiv 1706.03762 imported end to end (15 pages); bioRxiv bytes downloaded and passed the signature check, then Wordcell's Poppler parse failed identically to the anonymous path ("Poppler XML contains an invalid text top"); PMC returned its HTML bot challenge and produced the newno-accessmessage (the anonymous path reports only "does not have a valid PDF signature").🤖 Generated with Claude Code