qingflow-mcp 0.3.13 → 0.3.15

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (3) hide show
  1. package/README.md +9 -3
  2. package/dist/server.js +766 -245
  3. package/package.json +1 -1
package/README.md CHANGED
@@ -108,7 +108,7 @@ npm i -g git+https://github.com/853046310/qingflow-mcp.git
108
108
  Install from npm (pinned version):
109
109
 
110
110
  ```bash
111
- npm i -g qingflow-mcp@0.3.13
111
+ npm i -g qingflow-mcp@0.3.14
112
112
  ```
113
113
 
114
114
  Or one-click installer:
@@ -207,14 +207,20 @@ Deterministic read protocol (list/summary/aggregate):
207
207
  4. `strict_full=true` makes incomplete results fail fast with `NEED_MORE_DATA`.
208
208
  - for `qf_query(summary)`, `strict_full` enforces raw source scan completeness; sample rows may still be capped by `max_rows`, which is reflected by `output_page_complete=false`
209
209
  5. Error payloads expose `error_code` and `fix_hint` for actionable retries.
210
- 6. Parameter tolerance supports stringified JSON and numeric/boolean strings for key query fields.
210
+ 6. Public MCP `inputSchema` is strict:
211
+ - numbers must be native JSON numbers
212
+ - arrays must be native JSON arrays
213
+ - objects must be native JSON objects
214
+ - booleans must be native JSON booleans
215
+ - unknown fields are rejected by the MCP boundary
216
+ 7. Use `qf_query_plan` as the only preflight tool when the agent is unsure about arguments. It can normalize loose/model-shaped inputs before a real query is issued.
211
217
 
212
218
  For `qf_query(summary)` and `qf_records_aggregate`, read `data.summary.completeness` / `data.completeness` before concluding:
213
219
 
214
220
  1. `raw_scan_complete=false`: source data is not fully scanned, do not produce a final conclusion.
215
221
  2. `scan_limit_hit=true`: query stopped because scan budget was hit.
216
222
  3. `output_page_complete=false`: source may be complete, but output was truncated by `max_rows` or `max_groups`.
217
- 4. `raw_next_page_token`: use this token to continue raw scan pagination (`next_page_token` remains as a backward-compatible alias).
223
+ 4. `raw_next_page_token`: use this token to continue raw scan pagination (`next_page_token` remains as a backward-compatible alias). For `qf_query(summary)` / `qf_records_aggregate`, the token carries cumulative state, so keep query arguments unchanged when resuming.
218
224
 
219
225
  ## List Query Tips
220
226