iFinD

Use this skill to query iFinD via the local Python wrapper in scripts/ifind_api.py.

Environment preparation

Before first use, verify Python dependencies.

Required runtime dependency:

• requests

A pinned dependency file is provided at scripts/requirements.txt.

Check quickly with:

python3 -c "import requests; print(requests.__version__)"

If missing, install it before running any API script:

python3 -m pip install -r scripts/requirements.txt

If the environment is externally managed, prefer a virtual environment, but installing requests is allowed for this skill when needed.

First rule: confirm refresh_token before querying

Before any data call:

1. Check token status with:

python3 scripts/ifind_token_store.py status

1. If missing, prefer letting the agent obtain it directly from the official site:
• open https://quantapi.51ifind.com
• log in with the user's existing session, or ask the user to complete login if interactive approval is needed
• go to the account information page
• locate the refresh_token
2. Store it only through:

python3 scripts/ifind_token_store.py set --token ''

1. Only fall back to asking the user to provide the token if the browser path is unavailable or login cannot be completed.

Never ask the user to paste shell commands that would echo the token back into chat. Never print the token after storing it.

Preferred token acquisition workflow

Prefer this order:

1. Browser login to https://quantapi.51ifind.com and read the token from account information
2. iFinD 超级命令客户端 → 工具 → refresh_token 查询
3. Ask the user directly only as a fallback

If browser automation is used, keep the flow minimal and do not browse unrelated pages. Once the token is found, store it immediately and stop exposing it in further output.

Token storage policy

• Storage path: ~/.openclaw/skills/ifind/credentials.json
• File permission target: owner read/write only (600 on POSIX)
• The store script writes the file and tightens permissions automatically.
• The request script reads the refresh_token from that file unless IFIND_REFRESH_TOKEN is already present in the environment.
• Prefer the stored token over hardcoding tokens into code or notes.

Quick workflow

1. Verify dependencies and token state

Run:

python3 -c "import requests; print(requests.__version__)"
python3 scripts/ifind_token_store.py status

If requests is missing, install it first:

python3 -m pip install -r scripts/requirements.txt

If the token is missing, obtain it through the preferred browser workflow and store it first.

2. Choose a calling path

• For direct endpoint calls, use scripts/ifind_request.py endpoint
• For common market-data work, use scripts/ifind_request.py preset
• For Python integration or patching, read scripts/ifind_api.py
• For endpoint details and payload shape, read references/iFind_API_Reference.md

3. Run the query

Examples:

python3 scripts/ifind_request.py preset realtime \
  --codes '300033.SZ,600000.SH' \
  --indicators 'open,high,low,latest,changeRatio,volume,amount'

python3 scripts/ifind_request.py preset ohlcv \
  --codes '000300.SH' \
  --startdate '2025-01-01' \
  --enddate '2025-03-01'

python3 scripts/ifind_request.py endpoint get_thscode \
  --payload '{"seccode":"300033","functionpara":{"mode":"seccode","sectype":"","market":"","tradestatus":"0","isexact":"0"}}'

Output discipline

When using returned data in analysis:

• state the endpoint used
• include the symbol universe / code list
• include the requested date range or timestamp
• include the returned perf, dataVol, or other useful metadata when relevant
• if the API returns an error, surface the error code and message rather than guessing

References

• references/iFind_API_Reference.md: local API wrapper reference and examples from the user
• references/token-and-storage.md: refresh_token acquisition and storage procedure
• scripts/ifind_api.py: local Python wrapper
• scripts/ifind_request.py: CLI for endpoint/preset calls
• scripts/ifind_token_store.py: token status/set/remove helper