ntanh1402/sdlc-skills

sdlc-convert-doc

Convert local documents into Markdown by routing text-bearing files through Microsoft MarkItDown and OCR-heavy inputs through Datalab Marker.

Quelltext ansehen

Verwendung

  1. Installationsbefehl kopieren und das Quellpaket vor der Installation in der eigenen Agentenumgebung prüfen.
  2. Benötigte Werkzeuge, Zugangsdaten und Hosts im Original prüfen.
  3. Eingabe und gewünschtes Ergebnis beschreiben und die Ausgabe mit den Quellen abgleichen.
npx skills add https://github.com/ntanh1402/sdlc-skills --skill sdlc-convert-doc

Vor dem Start

  • Die Installation umfasst weder Modellabo noch API-Kontingent oder Dateizugriffsrechte.
  • Sterne und Installationen zeigen Aktivität, keine unabhängige Qualitäts- oder Sicherheitsprüfung.
Originales Skill-Dokument

Aus dem Quell-Repository gerendert; Überschriften, Beispiele, Code, Tabellen, Links und Bilder bleiben erhalten.

SDLC Doc to Markdown

Convert local files with the bundled router. Prefer MarkItDown for lightweight structured extraction and Marker when OCR or document-layout recovery is needed.

Workflow

Resolve SKILL_DIR to the directory containing this SKILL.md; do not assume the current working directory is the skill directory.

  1. Confirm the input is a local file and choose a separate .md output path.
  2. Run the dependency preflight:
bash
   python3 "$SKILL_DIR/scripts/check_dependencies.py" --engine auto
  1. Preview the routing decision when the file type or OCR need is uncertain:
bash
   python3 "$SKILL_DIR/scripts/convert_to_markdown.py" INPUT \
     --output OUTPUT.md --dry-run
  1. Convert with automatic routing:
bash
   python3 "$SKILL_DIR/scripts/convert_to_markdown.py" INPUT --output OUTPUT.md
  1. Read the generated Markdown. Check headings, reading order, tables, equations,

image links, missing text, and OCR corruption before reporting success.

  1. If a digital PDF produces sparse text, let the script retry it with Marker and

forced OCR. If quality is still poor, rerun explicitly:

bash
   python3 "$SKILL_DIR/scripts/convert_to_markdown.py" INPUT.pdf --output OUTPUT.md \
     --engine marker --force-ocr --overwrite

Never overwrite an existing output unless the user approved replacement. The script enforces this unless --overwrite is passed.

Engine selection

Input or conditionEngineReason
DOCX, PPTX, XLSX/XLS, HTML, EPUB, email, CSV, JSON, XML, ZIP, and ordinary text-bearing filesMarkItDownFast, lightweight structured extraction
PNG, JPEG, TIFF, BMP, GIF, or WebPMarkerImage inputs require OCR and layout recovery
PDF with usable embedded textMarkItDown firstAvoid unnecessary model loading
Scanned, image-only, sparse, or garbled PDFMarker with OCRRecover text and reading order
Layout-heavy PDF, equations, or complex tablesMarkerStronger layout and math handling

Use --engine markitdown or --engine marker only to override this policy. Use --page-range, --force-ocr, --strip-existing-ocr, --use-llm, or --disable-image-extraction for Marker-specific work. Marker-specific options make automatic routing choose Marker directly.

Output rules

  • Keep the source file unchanged.
  • Preserve Marker-generated images beside the Markdown so relative links remain valid.
  • Treat conversion as extraction, not proof of correctness. Compare critical values,

names, dates, tables, and formulas against the source.

  • Do not enable --use-llm without confirming the configured service and whether the

document may be sent to it.

  • Do not process an untrusted path or URL. The router accepts local files only.

Dependencies and troubleshooting

Require markitdown[all]==0.1.6 and marker-pdf==1.10.2; treat a missing or mismatched version as a failed preflight. Do not upgrade these pins implicitly.

Read references/converters.md when installing converters, choosing optional packages, reviewing supported formats, handling model downloads, or evaluating Marker licensing and compute requirements.

Run python3 "$SKILL_DIR/scripts/convert_to_markdown.py" --help for all conversion options.

aus demselben Repository

Weitere Skills

Alle Skills
ntanh1402
Community

sdlc-ask-wiki

Answer questions from the project wiki without changing it - list an application's Features with status, spec version, owner, priority, design readiness and open ChangeRequests; list what was written from an older version of a spec; show one Feature's approved architecture with its diagrams, services, frontends, contracts, decisions, traceability and what is not built yet; show one Endpoint or consumer with its input, output, validations and diagrams; show a user story or print its paste-ready c

Installationen / 1h
2
GitHub Stars
1
Aktualisiert
13. Nov.
ntanh1402
Community

sdlc-build-task

Write the code and unit tests of one wiki Task in one code repository, against the project wiki - read the Task's acceptance and planned scope, the contracts of the Design files it covers and the Application's conventions, plan the files, write them in a git worktree on a branch of their own, check them against the wiki and commit with a Task trailer, without pushing. Use when a developer asks to build, implement or code a Task of the project wiki. Not for coding without a wiki Task, for changin

Installationen / 1h
2
GitHub Stars
1
Aktualisiert
13. Nov.
ntanh1402
Community

sdlc-close-task

Record in the project wiki that the code of one wiki Task is merged - set the Task Done with its merge commit, remove the pending entries of the Design files it covered and mark them live, and move its Feature to InDev or its ChangeRequest to Implemented. Use after a Task's pull request is merged, when a developer asks to close, finish or complete a Task in the wiki. Not for closing a ticket in a tracker, for merging or reviewing code, for marking a Feature released, or for any other change of s

Installationen / 1h
2
GitHub Stars
1
Aktualisiert
13. Nov.
ntanh1402
Community

sdlc-design-arch

Design and approve the target architecture of a Feature or ChangeRequest in the project wiki, while its requirements are a draft or once they are agreed - the services, frontends, endpoints, events, data and externals it needs, their contracts, Mermaid diagrams, traceability and architecture decisions - and mark what is not built yet. Use when requirements need technical design, impact mapping, interface contracts or ADRs, when a spec changed after its design, or to approve a ChangeRequest whose

Installationen / 1h
2
GitHub Stars
1
Aktualisiert
13. Nov.