How to Configure the Obsidian Vault for the *patent-reader* Sub-Skill

Setting up the Obsidian vault for patent-reader requires two steps: register your vault path with check_obsidian_env.py, then run setup_obsidian_vault.py to bootstrap CSS snippets, templates, and index files.

The patent-reader sub-skill in the handsomestWei/patent-disclosure-skill repository generates structured patent-reading notes directly into an Obsidian vault. Proper configuration ensures seamless integration between the skill and your note-taking system. This guide walks through the exact commands and source code paths needed to configure the Obsidian vault for patent-reader.

Step 1: Register Your Vault Path

The skill must know where your Obsidian vault resides on disk. The helper script check_obsidian_env.py persists this path to the runtime configuration.

Run the script with the --set flag:

python skills/patent-reader/tools/vault/check_obsidian_env.py --set "D:/Obsidian/MyVault"

This writes to runtime_config()["obsidian_vault"] (see [shared/common.py](https://github.com/handsomestWei/patent-disclosure-skill/blob/main/shared/common.py)), which resolve_obsidian_vault() reads on every skill invocation. You only need to set this once; subsequent calls to the skill automatically resolve the vault.

Step 2: Bootstrap the Vault with setup_obsidian_vault.py

Once the path is registered, initialize the vault structure. The bootstrap script [setup_obsidian_vault.py](https://github.com/handsomestWei/patent-disclosure-skill/blob/main/skills/patent-reader/tools/vault/setup_obsidian_vault.py) performs three critical operations:

• Copies CSS snippets from assets/obsidian/ to .obsidian/snippets/
• Installs the base template (patents.base.yaml → patents.base) for consistent note formatting
• Creates the index note (_专利解读索引.md) with front-matter that the skill queries to discover related patents

Execute:

python skills/patent-reader/tools/vault/setup_obsidian_vault.py \
    --vault "D:/Obsidian/MyVault"

The script prints a summary of actions (lines 70–73) and lazily creates Research/Patents and Research/术语 folders if missing.

What Happens Inside bootstrap_vault()

The core logic lives in [obsidian.py](https://github.com/handsomestWei/patent-disclosure-skill/blob/main/skills/patent-reader/tools/vault/obsidian.py). Key functions include:

• repo_obsidian_base_src() — locates patents.base.yaml in assets/obsidian/
• bootstrap_vault() — orchestrates asset copying and index creation
• enrich_note_frontmatter() — adds read_date, tags, and aliases to generated notes

Optional: Manual Folder Structure

While setup_obsidian_vault.py creates folders on demand, you can pre-create them:

MyVault/
├── .obsidian/
│   └── snippets/          # CSS for patent note styling

├── Research/
│   ├── Patents/           # Patent reading notes: <pub>_解读_<date>.md

│   └── 术语/              # Glossary entries for technical terms

└── _专利解读索引.md        # Central index with YAML front-matter

Changing or Reconfiguring the Vault

To migrate to a new vault location, simply re-run the set command:

python skills/patent-reader/tools/vault/check_obsidian_env.py --set "E:/NewVault"

The next skill execution picks up the new path automatically. Re-run setup_obsidian_vault.py if the new vault lacks the required assets.

Code Reference: Resolving the Vault Path in Python

For custom integrations, import the resolution logic directly:

from shared.common import resolve_obsidian_vault, runtime_config

vault_path = resolve_obsidian_vault().get("vault") or runtime_config()["obsidian_vault"]
print(f"Obsidian vault: {vault_path}")

Creating Glossary Entries Programmatically

The [obsidian.py](https://github.com/handsomestWei/patent-disclosure-skill/blob/main/skills/patent-reader/tools/vault/obsidian.py) module exposes ensure_glossary_stub() for manual glossary management:

from pathlib import Path
from skills.patent_reader.tools.vault.obsidian import ensure_glossary_stub

vault = Path("C:/Obsidian/PatentVault")
rel_path, created = ensure_glossary_stub(
    vault,
    glossary_dir="Research/术语",
    term="光电传感器",
    definition="A sensor that converts light into an electrical signal.",
    source_pub="CN1234567",
)
print(f"Glossary page: {rel_path} (new={created})")

Key Source Files

File	Purpose
skills/patent-reader/tools/vault/check_obsidian_env.py	CLI helper to set/verify the vault path in runtime config
skills/patent-reader/tools/vault/setup_obsidian_vault.py	Bootstrap script that installs assets and creates index
skills/patent-reader/tools/vault/obsidian.py	Core library: vault resolution, front-matter handling, glossary utilities
skills/patent-reader/tools/vault/schema_vault.py	Canvas card schemas for visualization
shared/common.py	runtime_config() and resolve_obsidian_vault() implementations
assets/obsidian/	CSS snippets and patents.base.yaml template

Summary

• Store the vault path using check_obsidian_env.py --set so resolve_obsidian_vault() can locate it
• Bootstrap once with setup_obsidian_vault.py to install CSS, templates, and the index note
• Notes land in Research/Patents/ with enriched front-matter added by enrich_note_frontmatter()
• Glossaries live in Research/术语/ with optional programmatic creation via ensure_glossary_stub()

Frequently Asked Questions

What happens if I skip the bootstrap step?

The skill will fail when attempting to write notes. bootstrap_vault() in obsidian.py ensures .obsidian/snippets/, the base template, and _专利解读索引.md exist; without them, the skill cannot generate properly formatted output or maintain the index.

Can I use an existing Obsidian vault?

Yes. Point check_obsidian_env.py to any valid vault folder. The bootstrap script adds files non-destructively—it only creates missing directories and copies assets without overwriting existing notes.

Where does the skill read the vault path from?

resolve_obsidian_vault() checks runtime_config()["obsidian_vault"] first, which is populated by check_obsidian_env.py. This indirection allows the skill to run without hardcoded paths or environment variables.

How do glossary entries get created?

The skill automatically calls ensure_glossary_stub() when encountering technical terms during patent analysis. You can also invoke it manually (see code example above) to pre-populate definitions with specific source citations.

{
  "mcpServers": {
    "instagit": {
      "command": "npx",
      "args": ["-y", "instagit@latest"]
    }
  }
}