ClawHub

data-insight-generator

v1.0.0

零代码BI分析器，支持CSV/JSON数据自动清洗、指标计算、图表推荐与业务洞察输出

⭐ 0· 17·0 current·0 all-time
by@boboy-j
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
Name, description, schema, README, and SKILL.md all describe the same zero-code BI workflow (data quality checks, metric calculation, chart recommendations, insights). No unexpected binaries, env vars, or external services are requested. One minor provenance note: source/homepage is unknown which limits traceability but does not make the capability incoherent.
ℹ
Instruction Scope
SKILL.md confines runtime instructions to parsing the provided data_content and producing analysis/report output. It does not instruct reading other files, contacting external endpoints, or accessing environment variables. Practical privacy note: the skill expects users to paste data snippets (10–50 rows); SKILL.md/README advise masking sensitive fields, but the instructions do not enforce or automate redaction—so users could accidentally expose PII to the model.
✓
Install Mechanism
No install spec and no code files — instruction-only skill. This minimizes installation risk because nothing is written to disk or downloaded.
✓
Credentials
The skill requests no environment variables, credentials, or config paths which is appropriate for its stated purpose. The only input is user-supplied data_content; treat that as potentially sensitive.
✓
Persistence & Privilege
always is false and there is no behavior that modifies other skills or system settings. The skill can be invoked normally by the agent; that is expected and proportional.
Assessment
This skill is internally coherent and appears to do what it says: parse a pasted CSV/JSON snippet and produce a BI-style report. Before using it, do not paste full production data or raw PII (phone numbers, ID numbers, real names); mask or synthesize sensitive fields. Verify outputs before acting on suggested causal interpretations (the skill promises to distinguish correlation vs causation but that is inherently limited). Because the skill's source/homepage is unknown, prefer testing with non-sensitive examples first. If you need to analyze full datasets, prefer running a vetted local tool or an environment you control rather than pasting data into a third-party agent.

Like a lobster shell, security has layers — review code before you run it.

latestvk97e73c7nqxrcm91fwrst47tw5852p22

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Comments