Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Xiaodi Financial Team
v2.0.0金融分析师团队 - 股票分析、基金推荐、持仓复盘、投资决策
⭐ 0· 38·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
The name, description, SKILL.md instructions, and included agent prompts consistently describe a multi-role financial analysis team that fetches market and news data (Eastmoney, Tencent, Sina, Yahoo). These capabilities justify web_search/web_fetch/memory_search. However, _meta.json and parts of architecture.json also list the 'exec' tool and reference a script (stock_analysis.py) / scripts entry — but no scripts/code files are included in the package. 'exec' access (ability to run system commands) is not needed for the described read-only data-fetching and analysis tasks, making this an unexpected/extra privilege.
Instruction Scope
SKILL.md and agent prompts instruct fetching real-time market and news data and producing structured reports. Examples use curl to public finance endpoints (qt.gtimg, eastmoney APIs). The instructions do not ask for unrelated file/system access, secret exfiltration, or contacting unknown endpoints. They do reference a local configuration path ({baseDir}/config.json) which is reasonable for configuration.
Install Mechanism
No install spec or code files are provided — this is an instruction-only skill. That lowers install risk because nothing is downloaded or written during install. However, metadata implies a script and exec usage that would require files/execution if actually used; those files are absent.
Credentials
The skill does not request environment variables or external credentials (no API keys declared), which aligns with the public-data focus. The concern is metadata/tool requirements: listing 'exec' and scripts increases privilege expectations without justification. If the skill attempts to call private APIs later, it currently has no declared mechanism for storing credentials.
Persistence & Privilege
Flags are default (always:false, user-invocable:true) and the skill does not request persistent system-wide privileges or to modify other skills. Agent allowlists mention read/write/web_fetch which are reasonable for fetching and caching data and for report generation.
What to consider before installing
This skill appears to be a coherent financial-analysis assistant that uses public market and news APIs. Two things to verify before installing: (1) metadata and architecture reference the 'exec' tool and a script (stock_analysis.py) that are not included — ask the author why exec permission is declared and whether any code will be executed on your host (exec increases risk because it can run arbitrary commands). (2) Confirm the provenance: the package claims a GitHub/homepage; inspect that repository for the missing script(s) and any code that would run. If you don't need command execution, prefer a version that does not require 'exec' and that declares any external APIs/credentials explicitly. Consider running the skill in a sandboxed environment or reviewing the upstream repo before granting it elevated tool access.Like a lobster shell, security has layers — review code before you run it.
latestvk97d11gdxvntttrfx9v1r2f7gd83zj9k
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
📊 Clawdis