SKILLEMALL.ai
GitHub ↗ RU Sign in

BF majia-guanyuan

观远 BI（Guandata）实战增益层 Agent Skill —— 架在官方全家桶（guancli 查数 / guanvis 建卡发布截图 / guanetl ETL / guanwf 数据流 / guands 数据源 / guanmetric 指标写）之上，专攻官方 DSL/命令覆盖不到的硬骨头：Part B ETL 整库治理判断 + 10 类 BI 引擎报错手册 + SmartETL 全链路重写/ExecPlan、Part C 既有页自定义图表 HTML/CSS/JS 注入排障 + 固定卡/overlay、Part C-12 HTML 应用化看板（descriptor patch 把 selector 联到 custom chart 内部 dataView + 视觉设计底线/反 AI 味红线/guanvis screenshot 视觉验收）、Part D v7 草稿-发布状态机绕过 + SmartETL 节点化静默坑 + 移动端 phoneLayout ZIP inject、Part E SuperApp 开放应用反向工程（form 建表 /survey-engine/api/form/add + LLM 中转 ILLEGAL_JSON_RES 三路径解析 + 原生 fetch credentials 绕 unwrap）、AI-native ADS 数据架构方法论（治理 vs 重搭 / 7 字段约束 / 30+30+40 预算）、餐饮连锁 BI 公式实战库（60+ SQL/复购/RFM/AC/Comp/DWD 宽表范式/39 ETL 索引）。标准查数/建卡/ETL/数据集 CRUD 一律路由给官方全家桶，本 skill 专攻业务实战与引擎级踩坑。触发：营业额/门店/会员/订单/复购率/客单价/RFM/ETL 治理/payload_json/自定义图表/HTML 看板/应用化/看板太丑/AI 味/设计底线/观远/Guandata/v7 BI/60004 草稿/Spark 中文别名/customChart/phoneLayout/SuperApp/open-apps/form 建表/ILLEGAL_JSON_RES/AI-native ADS/数据架构重搭/DWD 宽表/洞察/Dashboard Agent/guanvis live/指标树/查询加速。

ClawHub Agent Skills author: 超级马甲 v3.1.10 MIT-0 38 files body ≈ 13 680 tokens Open the sourceclawhub.ai↗ analyzed 4 d ago

As a process F 31/100 · Will not run — References files that are not bundled: scripts/patch_selector_linkage.js

IntegrationData and analyticsAI and agentstype and topics are labelled automatically from the skill text
Runs in: OpenClaw Claude / Claude Code ChatGPT / Codex Gemini CLI Cursor GitHub Copilot / VS Code Qwen Code Kimi Code CLI Z.ai (GLM) DeepSeek (Deep Code) Mistral Vibe LangChain (deepagents) Goose OpenCode
JSON
Technical rating
B
76/100
safety, quality, tests
Safety 60%
95
Quality 40%
48
Run on models
none yet
Process rating
F
31/100
Will not run
References files that are not bundled: scripts/patch_selector_linkage.js
Tools and files w 18
0
Result and completion w 14
0
Inputs and preconditions w 11
0
the three weakest of ten parameters · all ten

What is at stake

Medium-severity findings: the skill is probably honest, but read what alarmed the scanner.

Dangerous commands medium severity

Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.

If you install

The skill contains commands that delete files, rewrite disks or execute code fetched from the network. The agent may run them without asking if it believes the instructions require it.

For the author

Replace destructive commands with safe equivalents that ask for confirmation, scope them to one folder, and stop piping curl into a shell: pin a version and a checksum.

How to improve

  1. Say in the description WHEN to use the skill ("use when…", example requests): that is the agent's main cue.
  2. The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
  3. The text references files that are not there: add them or drop the references.
For the model run — optional
  • Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
  • A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.

Guard findings · 1

✓ No critical or high findings

Medium and low: 1
  • medium Dangerous commands cmd-disable-security CHANGELOG.md:1159
    Disables a security control (Defender, Gatekeeper, SIP, firewall, execution policy) (quoted — discussed, not commanded)
    四步法。**重点收录 `xattr -dr com.apple.quarantine` 这一步**——2026-05-13
    quoted

Files scanned: 38. Evidence is masked. Grey chips explain why severity was lowered.

Against the Agent Skills spec

  • warning description-no-when description does not say WHEN to use the skill (no "use when")
  • warning body-long SKILL.md body ≈ 13680 tokens (recommended < 5000); move details to references/
  • warning missing-ref reference to a missing file: scripts/patch_selector_linkage.js

Process rating: all ten parameters 31/100

Will not run. References files that are not bundled: scripts/patch_selector_linkage.js
  • 0Tools and files. 1 referenced file(s) missing: scripts/patch_selector_linkage.js
  • 0Result and completion. Does not say what the result is
  • 0Inputs and preconditions. Does not say what the process needs to start
  • 20When it triggers. No condition that starts the skill
  • 30Running it twice. 72 mutating operations with no state check
  • 40Consistency. Frontmatter name (majia-guanyuan) differs from the folder (guanyuan-majia)
  • 40Execution cost. Instruction body is 13680 tokens: crowds the task out of the window
  • 50Failures and branches. 0 branches, has a failure section
  • 100Steps. 128 steps
  • 100Progress reporting. Reports progress
  • low 37 top-level sections: this looks like several domains in one skill
  • low The response is described with custom markup (18 tags): a typed call is more reliable

Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.

Quality signals

  • +5Description has no quoted example phrases that should trigger the skill
  • +4Description does not say when NOT to use the skill (false activations)
  • +3Description length 995: 120–800 characters recommended
  • +3Output format is not stated: the model decides each time
  • -5TODO / placeholder text left in the skill
  • -286 emoji in the instructions: noise for the model
  • -31 of 1 scripts are never mentioned in SKILL.md
  • +2Single-language instructions
  • +4Structure: 79 headings
  • +3Step-by-step instructions: 128 items
  • +4Has examples (31 code blocks)
  • +4Reference files are cited in the instructions (14 of 14)
  • +1License stated

Quality base 70; lint remarks subtract, signals add up to 100. Result: 48.

External checks

ClawHub: clean
This skill is a disclosed Guandata BI operations helper with powerful BI write/delete guidance, but its high-impact actions are purpose-aligned and gated by user-confirmation instructions.
LLM: benign (high) · VirusTotal: · 25 Aug 2026