doc-layout-parse
Prepare inputs for TeXada document checks. Use for a supplied layout.json tree, reviewed line-to-layout conversion via doc-ocr-extract, or routing images and PDFs through the local Studio transcription workflow. Distinguish reviewable page LaTeX from structured layout extraction.
Install
npx skills add https://github.com/CacinieP/TeXada-Studio-with-agentskills/tree/main/skills/doc-layout-parse
claude plugin marketplace add https://llmmart.ai/marketplace.json && claude plugin install caciniep-texada-studio-with-agentskills@llmmart
git clone https://github.com/CacinieP/TeXada-Studio-with-agentskills.git
The skills CLI installs just this skill, for any of its supported agents. Claude Code installs the whole caciniep/texada-studio-with-agentskills collection as a plugin from our marketplace. Git is the plain clone.
Skill manifest
Document Input Preparation
Choose the input route
The repository has three implemented routes. Select from the user's source and intended output:
| Source and goal | Route | Output |
|---|---|---|
| Existing document nodes for CLI formula/table checks | Inspect a supplied layout.json, then use the harness |
Node checks, candidates and a report |
Page image to CLI layout.json via reviewed OCR lines |
Run doc-ocr-extract (scripts/ocr_extract.py), review all candidate lines, then convert with scripts/ocr_to_layout.py |
Hash-bound layout.json (plus optional operator-annotated pixel regions) after explicit accept/reject decisions |
| PNG, JPEG, WebP or PDF to edit as LaTeX | Use Studio's image/PDF import with the configured local vision model | Page transcriptions merged into a reviewable .tex candidate |
Read input contracts and layout proposals for the relevant request and output fields. Do not treat the Studio .tex candidate as a layout.json tree.
Supplied layout tree
- Inspect examples under
samples/: a document hasdocandnodes; each node hasid,typeanddata. Formula data useslatex; table data usesrowsandexpected. - Keep
docunique within a run and node IDs unique within their document. Preserve existing page/source locations and crop paths; paths are resolved relative to the command's working directory. Sample crops are test doubles, not measured OCR evidence. - Run from the repository root when using repository-relative crop paths. The CLI accepts a directory containing
layout.json, or a parent with such directories one level below. It does not accept a PDF or generate nodes from a.texcandidate. - Resume only with stable input and execution context: terminal reuse is bound to their hashes, not just IDs. Use a fresh state directory for an independent evaluation.
Local Studio transcription
- Open the local Studio WebUI and use image/PDF import. Its authenticated
POST /api/studio/importcallsLocalRuntime.import_document; no separatescripts/parse.pyexists. - Import only the user's chosen source. The local route accepts a file up to 12 MiB, sends image content to an explicitly configured loopback vision endpoint, and handles at most five PDF pages per request. PDF rasterization requires
pdfinfoandpdftoppm. - Keep the original file and inspect the returned pages and LaTeX against it, including formulas, notation, page coverage and
[UNREADABLE]spans.requires_review: truemeans a candidate awaits review; the import does not save over an existing document. Creating a new Studio document is a separate explicit action. - Preserve the import ID, source/content hashes and available page metadata when the user needs provenance. Report the actual imported page range; a request starting after page 1 or returning
has_more_pages: truedoes not cover the whole PDF. BothPOST /api/studio/importand Studio's import UI supportpage_startalongside batch page limits (1,3,5), per-page exclusion or splitting into separate.texdocuments during review, and continuation batching whenhas_more_pages: true. - Compilation or a syntax check can help inspect the candidate, but neither establishes OCR fidelity or mathematical correctness. No automatic Studio-
.tex-to-layout.jsonconverter is implemented; for CLI checking afterward, prepare an explicitlayout.jsonfrom reviewed content or use the reviewed doc-ocr-extract +ocr_to_layout.pyline workflow on page images.
Scope and runtime loading
Studio implements page-to-LaTeX OCR, not evaluated block segmentation, bbox extraction, calibrated confidence or cross-page reading-order reconstruction. The reference's block-JSON prompt is an integration proposal, not the current API response. Never invent those fields or regenerate crop evidence by assumption.
This Skill guides an agent's workflow. Studio's OCR prompt is fixed in webui/local_runtime.py; it does not dynamically load this Skill into the model. The CLI formula host's separate Skill-loading trace must not be claimed as OCR Skill loading.
Guardrails
Keep inputs read-only. Write derived outputs to a separate state directory. Do not fetch private documents or send them to external services without authorization.
Files (texada-studio-with-agentskills)
-
evals
-
evals.json 822 B
Loading…
-
-
references
-
vlm-prompts.md 4.2 KB
Loading…
-
-
SKILL.md 4.6 KB
--- name: doc-layout-parse description: Prepare inputs for TeXada document checks. Use for a supplied layout.json tree, reviewed line-to-layout conversion via doc-ocr-extract, or routing images and PDFs through the local Studio transcription workflow. Distinguish reviewable page LaTeX from structured layout extraction. --- # Document Input Preparation ## Choose the input route The repository has three implemented routes. Select from the user's source and intended output: | Source and goal | Route | Output | | --- | --- | --- | | Existing document nodes for CLI formula/table checks | Inspect a supplied `layout.json`, then use the harness | Node checks, candidates and a report | | Page image to CLI `layout.json` via reviewed OCR lines | Run [doc-ocr-extract](../doc-ocr-extract/SKILL.md) (`scripts/ocr_extract.py`), review all candidate lines, then convert with [`scripts/ocr_to_layout.py`](../doc-ocr-extract/LAYOUT.md) | Hash-bound `layout.json` (plus optional operator-annotated pixel regions) after explicit accept/reject decisions | | PNG, JPEG, WebP or PDF to edit as LaTeX | Use Studio's image/PDF import with the configured local vision model | Page transcriptions merged into a reviewable `.tex` candidate | Read [input contracts and layout proposals](references/vlm-prompts.md) for the relevant request and output fields. Do not treat the Studio `.tex` candidate as a `layout.json` tree. ## Supplied layout tree - Inspect examples under `samples/`: a document has `doc` and `nodes`; each node has `id`, `type` and `data`. Formula data uses `latex`; table data uses `rows` and `expected`. - Keep `doc` unique within a run and node IDs unique within their document. Preserve existing page/source locations and crop paths; paths are resolved relative to the command's working directory. Sample crops are test doubles, not measured OCR evidence. - Run from the repository root when using repository-relative crop paths. The CLI accepts a directory containing `layout.json`, or a parent with such directories one level below. It does not accept a PDF or generate nodes from a `.tex` candidate. - Resume only with stable input and execution context: terminal reuse is bound to their hashes, not just IDs. Use a fresh state directory for an independent evaluation. ## Local Studio transcription - Open the local Studio WebUI and use image/PDF import. Its authenticated `POST /api/studio/import` calls `LocalRuntime.import_document`; no separate `scripts/parse.py` exists. - Import only the user's chosen source. The local route accepts a file up to 12 MiB, sends image content to an explicitly configured loopback vision endpoint, and handles at most five PDF pages per request. PDF rasterization requires `pdfinfo` and `pdftoppm`. - Keep the original file and inspect the returned pages and LaTeX against it, including formulas, notation, page coverage and `[UNREADABLE]` spans. `requires_review: true` means a candidate awaits review; the import does not save over an existing document. Creating a new Studio document is a separate explicit action. - Preserve the import ID, source/content hashes and available page metadata when the user needs provenance. Report the actual imported page range; a request starting after page 1 or returning `has_more_pages: true` does not cover the whole PDF. Both `POST /api/studio/import` and Studio's import UI support `page_start` alongside batch page limits (`1`, `3`, `5`), per-page exclusion or splitting into separate `.tex` documents during review, and continuation batching when `has_more_pages: true`. - Compilation or a syntax check can help inspect the candidate, but neither establishes OCR fidelity or mathematical correctness. No automatic Studio-`.tex`-to-`layout.json` converter is implemented; for CLI checking afterward, prepare an explicit `layout.json` from reviewed content or use the reviewed [doc-ocr-extract](../doc-ocr-extract/SKILL.md) + `ocr_to_layout.py` line workflow on page images. ## Scope and runtime loading Studio implements page-to-LaTeX OCR, not evaluated block segmentation, bbox extraction, calibrated confidence or cross-page reading-order reconstruction. The reference's block-JSON prompt is an integration proposal, not the current API response. Never invent those fields or regenerate crop evidence by assumption. This Skill guides an agent's workflow. Studio's OCR prompt is fixed in `webui/local_runtime.py`; it does not dynamically load this Skill into the model. The CLI formula host's separate Skill-loading trace must not be claimed as OCR Skill loading. ## Guardrails Keep inputs read-only. Write derived outputs to a separate state directory. Do not fetch private documents or send them to external services without authorization.
Comments (0)
Sign in to join the conversation.
Reviews (0)
No reviews yet.
No comments yet.