vitruvius-help
Skill10 starsChanged 4 days agoMITView on GitHub
Quick-reference card for all Vitruvius commands, disciplines, and the shared research method. One-shot display, not a persistent mode. Use when the user asks for Vitruvius commands, disciplines, or the shared research method. Trigger: /vitruvius-help, "vitruvius help", "what vitruvius commands", "how do I use vitruvius".
In this repository28 other files it gives its agents.
- AGENTS.mdAGENTS.md
- .github/copilot-instructions.mdCopilot instructions
- architecturalSkill
- artifact-readingSkill
- auditSkill
- civilSkill
- compareSkill
- design-alternativesSkill
- electricalSkill
- eli5Skill
- engineering-researchSkill
- evidence-rankingSkill
- fmea-brainstormSkill
- gap-analysisSkill
- habitSkill
- hypothesis-generationSkill
- mechanicalSkill
- peer-reviewSkill
- proposalSkill
- research-to-postSkill
- reviewSkill
- scholarly-researchSkill
- simplified-technical-englishSkill
- softwareSkill
- standards-lookupSkill
- summarizeSkill
- verifierSkill
- vitruviusSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 1 is distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
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 58
Every discipline skill runs the same loop (/skill:engineering-research):
- Distinctive: line 78
Never fabricate a source. A reference or it didn't happen.
- Very specific: line 79
Every output has a .provenance.md sidecar recording sources and verification status.
≈ “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
<!-- VITRUVIUS-COMPILED-SKILL:BEGIN v1 sha256=10aef9f57b1874050c1f090650753bdfcabad69d56cf10069e7da7b827d4b980 -->Vitruvius Help
Display this reference card when invoked. One-shot; do not persist anything.
What Vitruvius is
An engineering research agent: discover → read → synthesize → verify → review, with auditable provenance.
Commands
| Command | What it does |
/vitruvius <discipline or question> | Smart dispatcher — routes to discipline or skill automatically. |
/vitruvius:civil <question> | Civil / structural engineering research (buildings, bridges, steel, concrete, geotech). |
/vitruvius:mechanical <question> | Mechanical engineering research (design, thermal, fluids, materials). |
/vitruvius:software <question> | Software engineering research (architecture, frameworks, security). |
/vitruvius:electrical <question> | Electrical / electronics research (power, controls, EMC). |
/vitruvius:architectural <question> | Architectural research (building science, facades, codes). |
/vitruvius:verifier <claim> | Blind Verifier — independent subagent checks claim vs evidence. |
/vitruvius:compare <items> | Source/standard/design comparison matrix. |
/vitruvius:review <artifact> | Severity-graded artifact review. |
/vitruvius:audit <target> | Claim-vs-implementation mismatch audit. |
/vitruvius:summarize <doc> | Faithful structured document digest. |
/vitruvius:research-to-post <pdf-or-repo> | Draft 2–3 catchy hooks and a copy-first card with the best source image. |
/vitruvius:eli5 <topic> | Plain-language engineering explanation. |
/vitruvius:simplified-technical-english <text> | Rewrite technical text in Simplified Technical English (ASD-STE100). |
/vitruvius:artifact-reading <doc> | Anchored document reading + extraction. |
/vitruvius:scholarly-research <topic> | Academic literature discovery (OpenAlex, arXiv, etc.). |
/vitruvius:standards-lookup <standard> | Engineering standards (AISC, ACI, ASCE, IEEE, Eurocode). |
/vitruvius:gap-analysis <discipline> <topic> | Systematic literature gap identification via triangulation. |
/vitruvius:evidence-ranking <question> | Rank and score evidence by quality tier. |
/vitruvius:design-alternatives <problem> | Compare 3+ engineering approaches with scored trade-offs. |
/vitruvius:fmea-brainstorm <system> | Failure-mode brainstorming with S/O/D ratings and RPN. |
/vitruvius:proposal --posting X --cv Y | Targeted Ph.D./Masters proposal workflow. |
/vitruvius:habit | Extract research preferences for validation and approval. |
/vitruvius:engineering-research <question> | Shared research method (discover → read → verify → review). |
/vitruvius:hypothesis-generation <observation> | Freeze rival hypotheses with dated evidence boundaries. |
/vitruvius:peer-review <artifact> | Adversarial peer review with severity-graded findings. |
/vitruvius-help | This reference card. |
The shared research method
Every discipline skill runs the same loop (/skill:engineering-research):
- Plan — write
outputs/.plans/<slug>.md, get user confirmation. - Scale — direct search for narrow questions; researcher subagents only when decomposition clearly helps.
- Gather evidence — multiple query angles, primary sources first, an evidence table with stable numeric IDs.
- Draft — synthesize yourself, no invented sources or numbers.
- Cite — inline citations; verify every reference means what it claims.
- Review — adversarial verification pass; fix FATAL issues.
- Deliver — final file plus
.provenance.mdsidecar.
Seven canonical roles: researcher (gather), writer (synthesize),
verifier (cite + verify), reviewer (critique), arbiter (adjudicates
verifier disagreement), goal-checker (independent completion check), and
habit (read-only preference extraction). Subagents when the host supports
them, otherwise the lead agent performs the applicable role work.
Non-negotiables
- Never fabricate a source. A reference or it didn't happen.
- Every output has a
.provenance.mdsidecar recording sources and verification status. - Mark status honestly:
verified,inferred,blocked,unverified. - A numeric claim without a unit, sign convention, and source is noise — flag it.
Deactivate / stop
Say "stop vitruvius" or "normal mode" to stop an active research run. The skills remain available to invoke on demand.
Scope and Boundaries
- Vitruvius produces research — it does NOT produce final designs, construction documents, or implementation guidance.
- Research-only, not for final engineering sign-off. Licensed engineers must review and approve any design based on this research.
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.