extract_indicators
Pull multiple financial indicators for a Chinese company/year in one pass by fetching the annual report once, computing derived ratios, and caching results for reuse.
Instructions
Extract many indicators for one company/year in a single pass.
Fetches the annual-report PDF once, groups report-rules by section (one LLM call per section), dispatches Python extractors individually, and computes derived ratios locally. Caches the bundle to disk so repeat calls are free.
Tool call template
Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {
"ticker_or_name": "<ticker_or_name>",
"year": "<year>"
},
"name": "extract_indicators"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| form | No | form name; defaults to "年度报告". | 年度报告 |
| year | Yes | fiscal year. | |
| indicators | No | optional list of indicator names; omit to attempt every rule applicable to the company. Unknown / non-applicable names go to `missing`. | |
| extractor_mode | No | "auto" (default — each rule's declared extractor), "llm" (force LLM for report rules), or "python" (skip report rules whose extractor is LLM — LLM-free run). | auto |
| ticker_or_name | Yes | ticker or name (see get_company). |
Output Schema
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||