Skip to content

docs: add agent skill bundle for coding assistants (SKILL.md, pipelines, convert/evaluate)#3174

Open
jehlum11 wants to merge 5 commits intodocling-project:mainfrom
jehlum11:feat/agent-skill-docling-document-intelligence
Open

docs: add agent skill bundle for coding assistants (SKILL.md, pipelines, convert/evaluate)#3174
jehlum11 wants to merge 5 commits intodocling-project:mainfrom
jehlum11:feat/agent-skill-docling-document-intelligence

Conversation

@jehlum11
Copy link
Copy Markdown

Adds an Agent Skill-style bundle under docs/examples/agent_skill/docling-document-intelligence/ so tools like Cursor, Claude Code, and other assistants can follow a consistent convert → export JSON → evaluate → refine workflow when users ask to parse documents with Docling.
This is documentation + example scripts only; it does not change the docling library API or default behavior.

Contents

Item Purpose
README.md Overview, quick start, link to official docs / CLI
SKILL.md Full skill instructions (pipelines, chunking, evaluation loop); MIT frontmatter
pipelines.md Standard vs VLM pipelines, OCR engines, remote API notes
EXAMPLE.md Installing into ~/.cursor/skills/ and running helpers from the bundle root
improvement-log.md Optional template for local “what worked” notes
scripts/docling-convert.py CLI: Markdown, JSON, or hybrid chunks (standard / vlm-local / vlm-api)
scripts/docling-evaluate.py Heuristic quality report on export_to_dict JSON (+ optional Markdown path)
scripts/requirements.txt docling, docling-core, packaging (aligned with 2.81+)

Docs site

  • Linked from docs/examples/index.md
  • New nav entry in mkdocs.yml: Examples → Agent skill (Cursor / assistants)

Technical notes

  • Explicit PdfPipelineOptions wiring uses InputFormat.PDF + PdfFormatOption (required for docling 2.81+; string keys like "pdf" are invalid and error at runtime).
  • macOS OCR uses OcrMacOptions from the public API.
  • VLM API example includes generate_page_images=True and enable_remote_services=True where applicable.
  • Helper scripts are MIT (SPDX-License-Identifier: MIT) to match the repository.

How to verify locally

pip install -r docs/examples/agent_skill/docling-document-intelligence/scripts/requirements.txt
python3 docs/examples/agent_skill/docling-document-intelligence/scripts/docling-convert.py \
  https://arxiv.org/pdf/2408.09869 --out /tmp/docling-smoke.md
uv run mkdocs build   # from repo root with project env

- Add docs/examples/agent_skill/docling-document-intelligence/ with
  SKILL.md, pipelines.md, EXAMPLE.md, improvement-log template, and
  scripts/docling-convert.py + docling-evaluate.py (standard/vlm-local/vlm-api).
- Document InputFormat.PDF + PdfFormatOption for explicit PdfPipelineOptions.
- Link from examples index and mkdocs nav.

Made-with: Cursor
- Document both ~/.cursor/skills and docs/examples paths.
- README notes repo parity for PRs and local installs.

Made-with: Cursor
@github-actions
Copy link
Copy Markdown
Contributor

github-actions bot commented Mar 23, 2026

DCO Check Passed

Thanks @jehlum11, all your commits are properly signed off. 🎉

@mergify
Copy link
Copy Markdown

mergify bot commented Mar 23, 2026

Merge Protections

Your pull request matches the following merge protections and will not be merged until they are valid.

🟢 Enforce conventional commit

Wonderful, this rule succeeded.

Make sure that we follow https://www.conventionalcommits.org/en/v1.0.0/

  • title ~= ^(fix|feat|docs|style|refactor|perf|test|build|ci|chore|revert)(?:\(.+\))?(!)?:

I, jehlum11 <jehlum11@gmail.com>, hereby add my Signed-off-by to this commit: 2d268ff
I, jehlum11 <jehlum11@gmail.com>, hereby add my Signed-off-by to this commit: 041e709

Signed-off-by: jehlum11 <jehlum11@gmail.com>
Made-with: Cursor
@dolfim-ibm
Copy link
Copy Markdown
Member

@jehlum11 do you think we could instrument the SKILL for using the docling CLI? It looks like this is now adding a new script/cli for it. Is there any functionality missing in the CLI? In case we could add those.

@codecov
Copy link
Copy Markdown

codecov bot commented Mar 24, 2026

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

PeterStaar-IBM
PeterStaar-IBM previously approved these changes Mar 26, 2026
Address maintainer feedback: the custom docling-convert.py script was
largely redundant with the existing docling CLI. This commit:

- Removes scripts/docling-convert.py (redundant with `docling` CLI)
- Refactors SKILL.md (v1.4 → v2.0) to use `docling` CLI for all
  conversion tasks, reserving the Python API only for features the
  CLI does not expose (chunking, VLM API endpoint config,
  force_backend_text hybrid mode)
- Updates docling-evaluate.py recommended_actions to reference
  `docling` CLI flags instead of the removed script
- Updates README.md, EXAMPLE.md, pipelines.md to use `docling` CLI
  examples throughout
- Simplifies requirements.txt (removes packaging dependency)

The only custom script retained is docling-evaluate.py, which provides
heuristic quality evaluation — functionality the CLI does not cover.

Signed-off-by: jehlum11 <jehlum11@gmail.com>
Made-with: Cursor
dolfim-ibm
dolfim-ibm previously approved these changes Mar 30, 2026
Copy link
Copy Markdown
Member

@dolfim-ibm dolfim-ibm left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

lgtm

Signed-off-by: jehlum11 <jehlum11@gmail.com>
Made-with: Cursor
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants