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 agentmods add instructions/sunholo-data/ailang-parse/claude-mdgit clone --depth 1 https://github.com/sunholo-data/ailang-parseWrote 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/instructions/sunholo-data/ailang-parse/claude-md)<a href="https://agentmods.dev/instructions/sunholo-data/ailang-parse/claude-md"><img src="https://agentmods.dev/badge/instructions/sunholo-data/ailang-parse/claude-md.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00943 | $0.00943 |
| Opus 5 | $0.00472 | $0.00472 |
| Sonnet 5 | $0.00189 | $0.00189 |
| Haiku 4.5 | $0.00094 | $0.00094 |
Grade A, and why
ailang-parse CLAUDE.md 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 3d 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.
How it starts
The opening of the file, as written. The whole thing — 85 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md — AILANG Parse
Project Purpose
AILANG Parse is a standalone AILANG module for universal document parsing and generation. It extracts structured content from Office formats (DOCX, PPTX, XLSX, ODT, ODP, ODS, HTML, Markdown, CSV, EPUB, EML, TEX, RTF) deterministically and from PDFs/images via pluggable AI. LaTeX/arXiv parsing resolves \input/\include recursively with cycle detection, so multi-file papers (Vaswani, BERT, GPT-3) parse end-to-end. It also generates documents in 9 formats (including Quarto Markdown) from parsed content or AI prompts.
This is a production AILANG module, not a demo. Every change must exercise AILANG code paths.
Project Structure
ailang-parse/
├── docparse/ # AILANG modules (keeps docparse/ prefix for imports)
│ ├── types/document.ail # Block ADT (11 variants)
│ ├── services/ # Parser + generator modules
│ └── main.ail # CLI entry point
├── bin/docparse # Bash CLI wrapper
├── sdks/ # Python, JS, Go SDKs
├── data/test_files/ # Real-world test files
└── benchmarks/ # Benchmark infrastructure
Authoring documents in Markdown
Markdown is the input format you can WRITE, which makes it the way to build a test document with known-expected structure and then check what came out the other side. Front matter sets title/author; inline formatting, links, images (local paths are read and embedded), fenced code, blockquotes, nested lists and tables with alignment/colspan all survive into every output format.
It reaches roughly 70% of the generator surface. Headers, footers, comments and
tracked changes are NOT expressible in markdown — to exercise those, convert an
existing document that has them (e.g. data/test_files/docx-hdrftr.docx).
Generating a document and reading its structure back is how the last several defects were found; green suites did not see any of them.
AILANG Language & Toolchain Reference
Before writing or modifying AILANG code, load the full references:
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.
- 3d ago First seen · 85 lines · 943 tokens per session scan A 546ddc3b3bb4
ailang-parse CLAUDE.md is an instructions file published in the GitHub repository sunholo-data/ailang-parse (0 stars, last pushed 3d ago), licensed Apache-2.0. It adds 943 tokens to every session, about $0.0047 per session on Opus 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 2026-08-31.
Other instructions, from other repositories
mcp-ms-office-documents AGENTS.md
Instructions for ForLegalAI/mcp-ms-office-documents, covering agents.md, project overview, architecture, key conventions and filename generation.
mktero AGENTS.md
Instructions for tenglvjun/mktero, covering agents.md, project overview, commands, repository map and architecture and runtime rules.
go-docs-mcp CLAUDE.md
Instructions for drolosoft/go-docs-mcp, covering go-docs-mcp, build & run, architecture, key decisions and tools.
cdxml-toolkit-community CLAUDE.md
Claude Code instructions for ZiChenWang114514/cdxml-toolkit-community, covering cdxml-toolkit — agent reference, what this is, key rules, mcp server and tool reference.
office-open CLAUDE.md
Instructions for DemoMacro/office-open, covering project, architecture, build & test, measurement units and options api design.
tablepack AGENTS.md
Instructions for kujiangmudao/tablepack, covering two user paths, multimodal requirement, environment, hard requirements (never skip) and qc checklist (every pdf).