nenadzaric/skills-intel-liteparse repository overview

⁠liteparse

Parse the first 100 pages of one native-text PDF locally into JSON with text and bounding boxes. OCR is disabled; input is limited to 50 MiB and parsing to 120 seconds.

⁠Immutable image

docker pull --platform linux/amd64 docker.io/nenadzaric/skills-intel-liteparse@sha256:aa22ea0b933f16db983c531f7374917194c866cd5a743b4109fcc7afb5662c2b

Image index: sha256:d43ea7c62f1236d055125bf4b8ee3dadb799306187755983c7f1a06ba7940992.

⁠Source and license notices

This accompanying OCI artifact contains the original build recipe and runtime sources, pinned source directions, and original installed license/notices. Registry relocation and source-link metadata corrections in the current Git recipe do not change the image executable bytes. It is not a full mirror of every upstream source archive.

oras pull docker.io/nenadzaric/skills-intel-liteparse@sha256:797bd31d7c7efb72f72eec47bedbbb83e931a421fb23216f26d1439024274f6b
sha256sum liteparse-source-notices.tar.gz

Expected bundle SHA256: 3537a500aa29cfc2081390ee6b55bf5afb34d0dc6bc68005b71230439e687d79. Extract to read build-source/ and notices/SOURCE-AND-NOTICES.json.

Native license: Apache-2.0.

Pinned source archive: https://codeload.github.com/run-llama/liteparse/tar.gz/b754dc3ddc46f006e666b6e02ffc94d0937a5f3a⁠