Skip to content

fix: bound knowledge-base discovery and document extraction #117

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
mldangelo-oai wants to merge 22 commits into from
+2,981 −59
Closed

fix: bound knowledge-base discovery and document extraction #117

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
22 commits
Select commit Hold shift + click to select a range
c95cc9f
fix: bound knowledge-base ingestion
Jul 29, 2026
4e5f6fa
fix: bound PDF streams and reject mutable source escapes
mldangelo-oai Jul 30, 2026
3000c1f
Merge remote-tracking branch 'origin/main' into mdangelo/codex/bounde…
mldangelo-oai Jul 30, 2026
b688ed5
fix: recognize filesystem roots in knowledge-base containment
mldangelo-oai Jul 30, 2026
340e949
test: terminate simulated ChatGPT login deterministically
mldangelo-oai Jul 30, 2026
aaae7e1
fix: bound compressed PDF streams before parser allocation
mldangelo-oai Jul 30, 2026
3d6bc7b
fix: keep unusable DOCX numeric references as literal text
mldangelo Jul 30, 2026
a3f6dd7
fix: bound decoded document streams before model ingestion
mldangelo-oai Jul 30, 2026
1b64527
fix: bound every supported PDF compression pipeline
mldangelo-oai Jul 30, 2026
dab588e
fix: bound lexical PDF dictionary and LZW decoding work
mldangelo-oai Jul 30, 2026
2f33cb0
test: synchronously flush synthetic ChatGPT login prompts
mldangelo-oai Jul 30, 2026
a5433e7
fix: harden indirect PDF filters and decompression budgets
mldangelo-oai Jul 30, 2026
3ff3d4e
fix: support bounded modern PDF cross-reference streams
mldangelo-oai Jul 30, 2026
ee85980
test: prevent Node login fixture pipe hangs
mldangelo-oai Jul 30, 2026
35a7568
fix: safely bound PDF object and cross-reference parsing
mldangelo-oai Jul 30, 2026
5efe24a
fix: safely decode lexical PDF filters and image streams
mldangelo-oai Jul 30, 2026
dd6930f
fix: bound nested PDF filters, content arrays, and predictor work
mldangelo-oai Jul 30, 2026
139505d
fix: bound compressed PDF object discovery and stream indexing
mldangelo-oai Jul 30, 2026
15c4589
fix: decode every supported compressed PDF page dictionary
mldangelo-oai Jul 30, 2026
00223b9
fix: preserve complete bounded PDF object-stream semantics
mldangelo-oai Jul 30, 2026
602006e
fix: validate PDF stream structure and decoder parameters
mldangelo-oai Jul 30, 2026
9c6053b
fix: bound lexical PDF scans and resolve nested decode parameters
mldangelo-oai Jul 30, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 4 additions & 0 deletions sdk/typescript/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -223,6 +223,10 @@ to

Repeat `--knowledge-base PATH` for multiple files or directories. Directories are
searched recursively for Markdown, text, PDF, and Word (`.docx`) files.
Knowledge bases are limited to 128 documents and 4,096 discovered entries across
16 directory levels. Each input and extracted document is limited to 8 MiB,
aggregate input and extracted text are each limited to 32 MiB, and PDFs are
limited to 512 pages.

On macOS/Linux, an existing output directory must be private to the current
user (`chmod 700`).
Expand Down
Loading
Loading