dbs-standard-answer Skill for Claude: Theory and Historical Precedent for Hard Questions
Makes Claude answer a hard question with relevant theory and historical precedent, ending in a standard answer with stated conditions.
- Installs
- 13.3K
- on skills.sh
- Stars
- 11K
- on GitHub
- Files
- 1
- besides SKILL.md
- Updated
- Oct 2026
- last repo commit
dbs-standard-answer is one skill from the dbskill toolbox by dontbesilent. Given a hard question, Claude answers it first, then maps relevant fields, authors and theories and checks the premise. After that it compares historical successes, failures and counterexamples that share the same structure, and ends with a "standard answer" that states when it holds and when it fails.
SKILL.md and its output templates are written in Chinese. The skill bundles no scripts, only SKILL.md and an agents/openai.yaml file. It searches the web only when outside facts would change the conclusion or you ask for sources. It is not for finding a theory to back an existing PPT or method, which the toolbox handles in a different skill.
Good for
- Creators and teachers who want a theory behind a method, plus examples of it working and failing
- Founders and analysts asking whether anyone solved a problem like theirs before
- Anyone turning a theory into a short video, course or deck that helps real people
Install dbs-standard-answer
Claude Code
- 1
Copy this prompt.
Install the dbs-standard-answer skill in Claude Code. Run this command: claude plugin marketplace add dontbesilent2025/dbskill claude plugin install dbs-standard-answer@dontbesilent-skills Then check that it loaded and tell me how to use it.
- 2
Paste it into Claude Code. It shows the change and asks before running it.
Pick one route, not both. The README says plugin commands carry a plugin prefix (for the main package it is dbs:), so use the name Claude Code shows after install. The README's own one-liner npx -y skills add dontbesilent2025/dbskill -g --all installs all 34 skills into every agent the skills CLI finds, which this page does not recommend unless you want the whole toolbox.
From the project README. Source.
What's in the skill
When it loads
围绕复杂问题先挖掘相关行业、学科、作者与可信理论，再比较结构同构的历史成功、失败和反例，提炼带条件的标准答案。用户需要同时获得理论解释与历史比较，或明确询问历史上如何解决、有没有经典解法、标准答案时使用；只需给现有命题、案例、PPT 或方法论找理论依据时不使用。
Sections in SKILL.md
- 直接回答契约
- 共同起点
- Step 1：固定输入与任务
- Step 2：审查命题
- Step 3：提取底层机制
- 目标
- Step 1：建立跨领域知识地图
- Step 2：筛选并核实理论
- Step 3：人物与理论核实
- Step 4：把理论映射到用户的问题
- Step 5：理论解决大众问题
- Step 6：短视频应用
- 原型：{受众能听懂的问题标题}
- 第一部分默认输出
- 直接答案
- 第一部分：知识与理论挖掘
Key facts
- Installs on skills.sh
- 13,254 as of 2026-11-02
- GitHub stars
- 11,039 for the whole repo
- Last commit to the repo
- 2026-10-28
- Files in the skill
- 2 files: SKILL.md and agents/openai.yaml
- skills.sh audits
- Gen Agent Trust Hub pass, Socket pass, Snyk pass
What it includes
Answers your question first, then researches only as far as needed.
If you give no theory, Claude picks the mechanism itself instead of asking for one. It asks one question only when missing information would change the research target or mislead badly.
Try: “How do I turn this theory into a short video that solves real problems for more people?”
Read ·
Direct answer contractFix the input, review the claim, extract the underlying mechanism.
The claim review gives one verdict: keep, narrow, rebuild or drop. It checks for absolutes, causal jumps, switched definitions, over-generalizing and reverse inference.
Read ·
Common start (3 steps)Maps fields, authors and theories, verifies them, and applies one to your problem.
Six steps: knowledge map, theory scoring (below 7 of 10 cannot be the main theory), person and theory verification, mapping to your problem, problems the theory can solve, and short-video use.
Network ·
Part 1: knowledge and theory miningThree prototypes, each with a hook, a smallest action and a way to check the result.
Only when you ask for a short video. A prototype names the audience, the trigger scene, the cost of the old method and a step viewers can do today.
Write ·
Short-video prototypesTurns your problem into a structure and compares past cases.
Eight steps ending in a standard-answer grade: mature consensus, conditional answer, or no unified answer. Default is 4 to 6 cases covering closest success, cross-domain repeat, failure and counterexample.
Network ·
Part 2: historical isomorphismSearches only when needed; ranks sources from primary documents down to wikis.
Skips web search if you say not to use it. Two failed checks on a fact means it is marked unverified. Core historical facts should have two independent sources.
Network ·
Search and source standardsDirect answer, Part 1, Part 2, combined conclusion, sources.
Simple questions can be merged into a short answer. The skill then ends, and mentions /dbs only if it is installed.
Write ·
Final output order
From the official docs
Permissions
Low risk
The skill bundles no scripts. It can search the web, and its output includes historical claims, so check the sources it lists before you publish.
What it can do
- Read your data
- Create and change things
- Delete things
- Read-only mode
Watch out for
It may state historical claims you cannot check
The skill marks each fact as verified, interpretation, inference or unverified, and asks for sources. Open the sources for anything you will publish or teach. github.com
Technical details
| Scopes |
|
|---|
Questions
What does dbs-standard-answer do?
It makes Claude answer a hard question directly, then back the answer with relevant theories, authors and historical cases that share the same structure. It finishes with a conditional "standard answer" that states its limits and sources.
How do I install dbs-standard-answer in Claude Code?
Use the install box above. The plugin route installs only this skill. The README's --all command installs all 34 dbskill skills into every agent the skills CLI finds, so skip it unless you want the whole toolbox.
Does it need an API key?
SKILL.md names none. It uses web search only when outside facts would change the conclusion or you ask for sources, and it will not search if you tell it not to.
Is it free to use commercially?
Not by default. The repository's LICENSE and README say CC BY-NC 4.0, which allows non-commercial use with credit and asks you to contact the author for commercial use. GitHub itself does not detect a license for the repo.
Is dbs-standard-answer safe?
The skill bundles no scripts. It can search the web and may state historical claims, so check the sources it lists. SKILL.md labels each fact as verified, interpretation, inference or unverified.
dbs-standard-answer vs dbs-theory-grounding?
Pick dbs-standard-answer when you want theory plus historical comparison for a hard question. dbs-theory-grounding's SKILL.md describes a single theory-anchoring pass for a course, PPT, copy or method, and says it does not expand into historical research by default. dbs-standard-answer's SKILL.md says not to use it when you only need support for an existing proposition, case, PPT or method.
Information
updated Nov 2, 2026
- Repository
- dontbesilent2025/dbskill
- Folder
skills/dbs-standard-answer
Sources
Docs, issue trackers and discussions this page is built from.
Show all 6Hide
- dbs-standard-answer folder · github.com
- dbskill README · github.com
- skills CLI commands · github.com
- SKILL.md · github.com
- LICENSE · github.com
- dbs-theory-grounding SKILL.md · github.com