research-intake
Skill1 starsChanged 4 months agoMITView on GitHub
Formalize a new technical or academic research question in a RigorGraph project while separating goals, assumptions, evidence, and open gaps. Use when starting research, clarifying an ambiguous question, or defining completion criteria before proof, experiments, or literature work.
In this repository4 other files it gives its agents.
- AGENTS.mdAGENTS.md
- adversarial-verifySkill
- capture-claimSkill
- release-auditSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 2 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 8
Read rigorgraph.yaml and existing .rigorgraph/ records when present.
- Very specific: line 12
If a later CI or experiment run will produce a RigorGraph Evidence Bundle, record the expected scope and producer as a plan only.
- Distinctive: line 17
Use these localized headings:
- Distinctive: line 27
Never present model consensus, a finite scan, or an unverified citation as a verified result.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Research intake
- Read
rigorgraph.yamland existing.rigorgraph/records when present. - Determine the user's language from the explicit request, project language, or system language. Reply in that language: English, Traditional Chinese, Simplified Chinese, or Japanese.
- Clarify the domain, variables, quantifiers, assumptions, target claim type, allowed evidence, constraints, and completion test. Ask only for material missing information.
- Separate established facts, user assumptions, proposed claims, and open gaps. Do not attempt a proof or literature conclusion during intake.
- If a later CI or experiment run will produce a RigorGraph Evidence Bundle, record the expected scope and producer as a plan only. Do not fabricate the bundle, digest, provenance, or successful result during intake.
- Preserve mathematical notation and quoted source text exactly; never translate user-authored claims unless asked.
- Initialize the project with
rigorgraph initif needed. Record candidate claims only asDRAFTorPROPOSEDthroughrigorgraph claim add. - End with the formalized question, assumptions, completion criteria, open gaps, and the safest next step.
Use these localized headings:
| en | zh-TW | zh-CN | ja |
| Research question | 研究問題 | 研究问题 | 研究課題 |
| Assumptions | 假設 | 假设 | 仮定 |
| Completion criteria | 完成標準 | 完成标准 | 完了基準 |
| Open gaps | 未解缺口 | 未解缺口 | 未解決のギャップ |
| Next safe step | 下一個安全步驟 | 下一步安全操作 | 次の安全な手順 |
Never present model consensus, a finite scan, or an unverified citation as a verified result.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.