gitea-admin/obsidian-wiki
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

refactor(mcp): extract order services into independent project

Browse Source 
Remove Kingdee sync, business CLI/MCP tools, and internal order-query routing
now owned by the separate kingdee-orders-mcp project. Keep Wiki retrieval,
file assets and ingest available in code while its deployed service is off.
Retain legacy response fields with business access disabled, and document
migration ownership and rollback without exposing credentials or data.

Validation: 174 Wiki tests pass. Independent order service verified with
36 template parity checks, authenticated HTTP queries, and a real sync.
This commit is contained in:
Codex
2026-10-31 07:41:18 +00:00
parent 6c451c3eb6
commit 2396bd2d3e
25 changed files with 47 additions and 5793 deletions
Download Patch File Download Diff File Expand all files Collapse all files
AGENTS.md
+1 -1
View File
@@ -126,7 +126,7 @@ See `wiki-query` and `wiki-export` skills for how the filter is applied.
## File Asset Indexing
- **Keep business MCP independent.** The opt-in `business` group exposes direct schema/template/guarded-SQL/status tools without Wiki retrieval or OpenCode. `scripts/mcp_service.sh tools set business` disables Wiki query, file assets, and ingest together while preserving order access. Business MCP reads the existing SQLite database through read-only connections; the separate Kingdee sync task owns updates. Preserve configured internal-party exclusions and sync freshness metadata. `OBSIDIAN_MCP_BUSINESS_DB` or `--business-db` selects an existing business database independently of the file-asset index.
- **Keep order services separate.** Order MCP, Kingdee synchronization, credentials, and the live order database belong to the independent `kingdee-orders-mcp` project. Wiki must not import its code or read the legacy order database. The old database is retained only for migration rollback. Client agents may combine results from the two independent MCP services.
- **Index source files as richly as possible.** When writing `_meta/source-registry.jsonl` or rebuilding `_meta/file-assets.sqlite`, preserve structured identifiers instead of relying only on filenames or summaries.
- **Keep product lookup fields explicit.** Carry forward `bjd_part_numbers`, `competitor_part_numbers`, `series`, `pitch`, `pin_count`, `product_lines`, `customers`, `competitors`, `standards`, `business_categories`, `related_pages`, and `catalog_paths` whenever they are known.