BD sn-deep-research
用于用户请求深度研究、系统性研究、竞品分析、方案对比、趋势分析或事实核查时。**遇到以下任一情况就主动使用本 skill，不要自行搜几条就回答**：①用户出现触发词：深度研究 / 深度调研 / 深入研究 / 全面研究 / 系统研究 / 调研 / 调查 / 尽调 / 行业研究 / 市场研究 / 竞品分析 / 政策研究 / 技术研究 / 趋势研究 / 事实核查 / 写一份研究报告 / 调研报告 / 深度报告 / research / deep research；②请求需要跨多来源取证、多维度对比、交叉验证才能给出可靠结论；③用户要求产出报告、白皮书、行业分析或尽调文档；④话题涉及最新政策/市场/产品/价格/法规，需要系统核查。无核验要求的简单常识问答不使用。模糊或宽泛的"研究/了解一下 X"也优先触发。仅不用于：一句话摘要、已给定单一来源的整理、纯文字润色改写。
用于用户请求深度研究、系统性研究、竞品分析、方案对比、趋势分析或事实核查时。遇到以下任一情况就主动使用本 skill，不要自行搜几条就回答：①用户出现触发词：深度研究 / 深度调研 / 深入研究 / 全面研究 / 系统研究 / 调研 / 调查 / 尽调 / 行业研究 / 市场研究 / 竞品分析 / 政策研究 /…
As a process D 44/100 · Unfinished process — weak spots: result and completion, when it triggers, inputs and preconditions
What is at stake
Medium-severity findings: the skill is probably honest, but read what alarmed the scanner.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The instructions or scripts send data out: environment variables, keys, file contents, chat history. You may never notice, because the agent performs the upload as "part of the task".
If the upload is not needed for the task, remove it: catalogs flag such skills and delist them. If it is needed, name the destination explicitly, say what leaves the machine, and give the user a switch.
How to improve
- Say in the description WHEN to use the skill ("use when…", example requests): that is the agent's main cue.
- The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
- 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 · 2
✓ No critical or high findings
Medium and low: 2
-
medium Exfiltration
net-redirectable-api-keyworkbench-runtime/bin/sensenova-ppt-workbench.mjs:489Helper sends the API key to a host configured by an environment variable — the key can be redirected to another serverAPI key + configurable base URL from environment
-
low Obfuscation
uni-mixed-script-wordworkbench-runtime/dist/assets/MarkdownArtifactViewer-C7xIspsK.js:2Word mixing Latin and Cyrillic letters (homoglyph obfuscation) (33 occurrences) (detector / deny-list definition)`,` `],T=[`/`,`?`,`#`],E=255,D=/^[+a-z0-9A-Z_-]{0,63}$/,re=/^([+a-z0-9A-Z_-]{0,63})(.*)$/,O={javascript:!0,"javascript:":!0},ie={http:!0,https:!0,ftp:!0,gopher:…detector
Files scanned: 27. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
description-no-whendescription does not say WHEN to use the skill (no "use when") - warning
body-longSKILL.md body ≈ 5072 tokens (recommended < 5000); move details to references/
Process rating: all ten parameters 44/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Failures and branches. Linear process with no failure handling
- 0Progress reporting. Says nothing while it works
- 20When it triggers. No condition that starts the skill
- 60Tools and files. Uses tools (bash) that frontmatter does not declare
- 70Execution cost. Instruction body is 5072 tokens
- 100Steps. 63 steps
- 100Consistency. Name and required fields are in place
- 100Running it twice. No mutating operations
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)
- +3Output format is not stated: the model decides each time
- -32 of 5 scripts are never mentioned in SKILL.md
- +1No license
- +2Single-language instructions
- +3Description length 384: enough signal without eating the budget
- +4Structure: 28 headings
- +3Step-by-step instructions: 63 items
- +4Has examples (17 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 59.