股智Alpha
ReviewAudited by ClawScan on May 12, 2026.
Overview
The skill appears to be a stock-analysis tool, but it needs review because its artifacts can overstate the data source and scan scope behind financial recommendations.
Before installing, verify the actual command to run, use an isolated Python environment, and treat reports as informational rather than investment advice. Pay special attention to whether fund-flow data is real akshare data or an estimate, and whether a scan covered the full market or only a core pool.
Findings (5)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
A user could treat estimated fund-flow and SLSV scores as if they came from a real main-capital-flow data source when making investment decisions.
The single-stock report labels any non-empty fund-flow result as 'akshare'. In the provided FundFlowCollector, the default path can estimate fund flow from price/volume data rather than using akshare, while SKILL.md presents fund flow as real-time akshare data.
资金流: {'价量估算' if len(flow_df)==0 else 'akshare'}Either actually use akshare when claiming akshare data, or clearly label estimated fund-flow data in SKILL.md and every report output.
A user may believe the Top N list is selected from the whole A-share market when it may only be selected from a smaller default pool.
The default screener uses a core stock pool unless --full-scan is supplied, while the skill description and examples emphasize A-share full-market scanning.
codes = args.codes.split(",") if args.codes else get_pool(core_only=True)Make the default scan scope explicit in user-facing output and SKILL.md, and document or use --full-scan whenever claiming full-market coverage.
The agent may not invoke the intended screener reliably.
SKILL.md repeatedly documents main.py as the entrypoint, but the provided file manifest does not include main.py. If followed from the wrong working directory, this could fail or run an unintended local file named main.py.
python main.py --mode scan --top-n 10
Package the referenced main.py or update SKILL.md to call the included script, such as python scripts/run_screener.py, using a clear skill-relative path.
Installation may pull newer dependency versions than the publisher tested.
The skill depends on PyPI packages using lower-bound version ranges rather than pinned versions or a lockfile. This is common for Python tools but means future dependency changes can alter behavior.
pandas>=2.0.0 numpy>=1.24.0 httpx>=0.24.0 akshare>=1.18.0
Install in an isolated virtual environment and prefer pinned, reviewed dependency versions for reproducible behavior.
Third-party providers can see queried stock symbols, and tampered or unavailable data could affect the report.
The skill retrieves market data from external HTTP endpoints. This is expected for a stock-data skill, but the data source and network path influence the recommendations.
URL=f'http://money.finance.sina.com.cn/quotes_service/api/json_v2.php/CN_MarketData.getKLineData?symbol={code}&scale={ts}&ma=5&datalen={count}'Treat outputs as informational, verify important data against trusted sources, and prefer HTTPS/data-source validation where possible.