Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add airmang/python-hwpx-automation --skill reference-preserving-editgit clone --depth 1 https://github.com/airmang/python-hwpx-automationWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/airmang/python-hwpx-automation/reference-preserving-edit)<a href="https://agentmods.dev/skills/airmang/python-hwpx-automation/reference-preserving-edit"><img src="https://agentmods.dev/badge/skills/airmang/python-hwpx-automation/reference-preserving-edit/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/airmang/python-hwpx-automation/reference-preserving-edit"><img src="https://agentmods.dev/badge/skills/airmang/python-hwpx-automation/reference-preserving-edit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00046 | $0.00436 |
| Opus 5.5 | $0.00018 | $0.00174 |
| Sonnet 5.5 | $0.00009 | $0.00087 |
| Haiku 4.5 | $0.00005 | $0.00044 |
Grade A, and why
reference-preserving-edit scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Reference-Preserving Edit
Use this workflow when the document already exists and layout stability matters more than editing speed.
Tool Order
- Create a working copy with
copy_document. - Read the current structure with
get_document_outline,get_paragraphs_text,get_table_text, orhwpx_extract_json. - In advanced mode, inspect sensitive package parts with
package_parts,package_get_xml,package_get_text,object_find_by_tag, orobject_find_by_attr. - Validate the baseline with
validate_structureif the source document is externally produced or suspicious. - Apply the smallest possible edit:
search_and_replaceorbatch_replacefor anchored text updatesformat_textfor localized stylingset_table_cell_textfor fixed cells- paragraph insertion/deletion only when structure must change
- Re-read the touched areas and rerun
validate_structure.
Minimize Layout Drift
- Prefer replacement over insertion.
- Avoid changing table geometry unless required.
- Limit edits to the smallest paragraph or table region possible.
- Use package inspection when memo anchors, section XML, or dense table layouts make visual assumptions risky.
When To Inspect Package Parts
- The source was generated by another system.
- The visible text is not enough to identify the true anchor.
- You need to confirm which section or package part contains the target content.
Honest Limitations
apply_document_commandssupports a typed set/add/remove/move/copy union, not arbitrary package mutation; use a domain-specific tool when its evidence or semantics are required.- There is no public layout-drift report or structure-diff tool on the active MCP surface.
- Saving is implicit: mutating tools persist immediately, so always start from
copy_documentwhen the original must stay untouched.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 41 lines · 46 tokens per session scan A 418448953a19
reference-preserving-edit is a skill published in the GitHub repository airmang/python-hwpx-automation (68 stars, last pushed 4d ago), licensed Apache-2.0. It adds 46 tokens to every session and 436 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
hwpx
A tool for reading, editing, filling, and creating Korean Hancom document files. It handles `.hwpx` documents, which are ZIP-based XML files, and `.hwp` documents, an older binary format.
kordoc
A command-line and server tool for reading and creating Korean official documents, including HWP, HWPX, PDF, DOCX, and spreadsheet files. It can convert documents to Markdown and create or update HWPX files.
hwpx
A toolset for reading, editing, filling, and generating Korean HWPX documents. HWPX is a ZIP-based document format used by Hancom Hangul and follows the OWPML document structure.
obsidian-vault-workflow
Use this skill when a user asks an agent to inspect, organize, summarize, or update an Obsidian vault through this MCP server.
ai-hwp-reader
A read-only tool for opening Korean HWP and HWPX word-processing files, or ZIP archives containing them, without Hancom Office. It converts the contents to Markdown while preserving document structure.
ai-hwp-reader
An AI skill that tells a coding agent how to read Korean HWP and HWPX word-processing files, including files inside ZIP archives. It extracts the content and then uses it to complete the user's requested task.