← Back to Skills Marketplace
ivangdavila

Engineering

by Iván · GitHub ↗ · v1.0.0
linuxdarwinwin32 ✓ Security Clean
1583
Downloads
2
Stars
8
Active Installs
1
Versions
Install in OpenClaw
/install engineering
Description
Support engineering understanding from DIY projects to professional practice and research.
README (SKILL.md)

Detect Level, Adapt Everything

  • Context reveals level: vocabulary, technical depth, professional credentials
  • When unclear, ask about their role before giving specific guidance
  • Always state safety factors, units, and assumptions explicitly

For Hobbyists: Accessible Without Dumbing Down

  • Explain the "why" behind calculations — "Wood grain direction affects strength; here's how that changes your bracket design"
  • State safety margins explicitly — "Use 3/4" plywood minimum though 1/2" would theoretically hold; extra gives margin for knots and humidity"
  • Flag professional-required systems — electrical mains, load-bearing mods, gas lines, pressure vessels require permits and licensed review
  • Provide material alternatives with trade-offs — "6061-T6 aluminum is ideal but hard to source; 3mm steel flat bar is heavier but easier to drill"
  • Include tool-availability checks — "Best welded, but with drill and hacksaw, use bolted angle brackets with gusset plate"
  • Quantify forces in relatable terms — "200 lbs shear force means two adults standing on it; your 1/2" bolt handles 800 lbs, so 4x safety margin"
  • Identify failure modes and consequences — "If weld cracks, shelf drops suddenly. If wood splits, it gives warning creaks first. Design for gradual failure."
  • State when codes apply — "Deck railings have code requirements (42" height, baluster spacing, 200lb lateral). Follow them; people die from falls."

For Students: Principles and Rigor

  • Show complete problem-solving methodology — identify knowns/unknowns, draw diagrams, select equations, solve symbolically first, then substitute with units
  • Enforce unit consistency — verify units at every step; convert to consistent systems before computing; flag mismatches
  • Explain physical intuition — why relationships exist, what each term represents, what happens when variables change
  • Reference fundamental principles — state which law applies (Conservation of Energy, Newton's Laws, Kirchhoff's Laws) and why
  • Provide worked examples with increasing complexity — start idealized, progressively add friction, transients, nonlinearities
  • Connect theory to practical applications — cite real systems: engines for thermodynamics, trusses for statics, op-amps for electronics
  • Support derivations — be prepared to derive key equations from first principles
  • Identify common misconceptions — sign conventions, passive sign convention, reference frames, stress vs strain, power vs energy

For Professionals: Standards and Liability

  • Cite specific code versions and sections — "per ASME B31.3-2022 §304.1.2" not just "per code"; versions matter for liability
  • Flag jurisdiction amendments — remind to verify with Authority Having Jurisdiction (AHJ) for final compliance
  • Distinguish prescriptive from advisory — "shall" is mandatory; "should" is recommendation
  • Include safety factor assumptions — state what SF was used and why; "Using SF=4 per standard practice for lifting equipment"
  • Warn when operating near limits — if calculation shows 85%+ of allowable, flag as "low margin, verify assumptions"
  • Include PE review disclaimer — "This analysis must be reviewed and stamped by a licensed Professional Engineer before use"
  • Flag cross-discipline interfaces — "This touches structural/electrical/process; coordinate with licensed specialist"
  • Use discipline-standard terminology — default to industry conventions (psig vs psia); maintain consistent unit systems

For Researchers: Validation and Rigor

  • Enforce experimental design principles — proper controls, statistical power, uncertainty quantification
  • Distinguish simulation from validation — never accept simulation as proof; recommend validation hierarchy (component → subsystem → system)
  • Adhere to publication standards — know IEEE, ASME, Elsevier formatting; reference DOIs; flag predatory journals
  • Require quantified uncertainty — reject "good agreement" without confidence intervals and error bounds
  • Apply appropriate skepticism — distinguish peer-reviewed advances from hype; recommend landmark papers, not preprints
  • Prioritize reproducibility — encourage sharing datasets, code, CAD files, protocols; apply FAIR data principles
  • Match modeling fidelity to question — don't over-compute when simpler models suffice; don't oversimplify when physics demands resolution
  • Navigate interdisciplinary rigor — apply stricter standards of each field; don't let approximations bypass adjacent-science requirements

For Educators: Fundamentals and Practice

  • Build from first principles before formulas — establish underlying physics before introducing equations
  • Require unit analysis on every calculation — reject answers without units; catches 70%+ of errors
  • Scaffold idealized to real-world — start simplified (frictionless, steady-state), add complexity progressively
  • Actively probe misconceptions — force vs pressure, sign conventions, vectors as scalars, linear assumptions in nonlinear systems
  • Connect to codes and standards — reference AISC, NEC, ASME; real engineering requires compliance
  • Emphasize estimation before calculation — sanity-check answers; engineers who can't estimate are dangerous
  • Require diagrams before calculation — FBDs, control volumes, circuit diagrams; no diagram means no solution attempt
  • Simulate exam conditions — provide problems in PE/FE exam format with time pressure and ethics scenarios

For Technicians: Implementation and Escalation

  • Reference specific drawing callouts — cite sheet number, detail reference, revision letter, date; never assume "current drawing"
  • Provide step-by-step troubleshooting — numbered procedures with expected readings; decision trees for branches
  • State tolerances and calibration — specify acceptable ranges, instrument accuracy class, calibration requirements
  • Distinguish scope clearly — flag when PE review required for modifications, recalculations, design changes
  • Cite codes by section — exact sections with edition year for compliance documentation
  • Provide verification checklists — quantitative pass/fail criteria (torque values, clearances, test hold times) for QA documentation
  • Document as-built discrepancies — specify deviation, whether within variance, proper RFI process if engineering review needed
  • Include safety protocols — LOTO requirements, minimum PPE, confined space protocols for any hands-on procedures

Always

  • State assumptions, safety factors, and units explicitly
  • Distinguish theory from validated practice
  • Flag when professional review or permits are required
  • Engineering errors can kill; err on the side of safety
Usage Guidance
This skill is coherent and instruction-only (nothing is installed and it asks for no secrets). Before relying on it for safety-critical designs: 1) treat its calculations as advisory — get a licensed Professional Engineer (PE) review where required; 2) verify units, assumptions, and boundary conditions it states; 3) provide clear role/context, drawings, and required constraints so responses stay accurate; and 4) if you do not want autonomous agent invocation, consider disabling the skill or restricting when it can run (the platform default allows autonomous invocation). If you need offline or jurisdiction-specific compliance checks, do those with the appropriate human experts and official code references.
Capability Analysis
Type: OpenClaw Skill Name: engineering Version: 1.0.0 The skill bundle contains standard metadata and a SKILL.md file with detailed instructions for an AI agent to provide engineering support. The instructions are entirely focused on improving the quality, safety, and rigor of the engineering advice, tailored for different user levels. There are no indications of prompt injection attempts, unauthorized command execution, data exfiltration, or any other malicious behavior. All content aligns with the stated purpose of an 'Engineering' skill.
Capability Assessment
Purpose & Capability
The name/description (general engineering guidance across hobbyist → professional → research) matches the SKILL.md content: level-adaptive advice, safety factors, code references, and pedagogy. The skill requests no binaries, env vars, or config paths — appropriate for an advice/analysis skill.
Instruction Scope
The SKILL.md instructs the agent to ask about user role, state assumptions/units, show calculations, request diagrams/drawings when needed, and include disclaimers for professional review. It does not instruct reading system files, accessing environment variables, or sending data to external endpoints beyond normal conversation flow.
Install Mechanism
No install spec and no code files are present. This lowers risk: nothing will be written to disk or fetched at install time.
Credentials
No environment variables, credentials, or config paths are required. The SKILL.md does not reference hidden credentials or unrelated services.
Persistence & Privilege
The skill is not marked always:true and uses default model-invocation behavior. It does not request persistent system-level privileges or modify other skills/configuration.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install engineering
  3. After installation, invoke the skill by name or use /engineering
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release
Metadata
Slug engineering
Version 1.0.0
License
All-time Installs 8
Active Installs 8
Total Versions 1
Frequently Asked Questions

What is Engineering?

Support engineering understanding from DIY projects to professional practice and research. It is an AI Agent Skill for Claude Code / OpenClaw, with 1583 downloads so far.

How do I install Engineering?

Run "/install engineering" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Engineering free?

Yes, Engineering is completely free (open-source). You can download, install and use it at no cost.

Which platforms does Engineering support?

Engineering is cross-platform and runs anywhere OpenClaw / Claude Code is available (linux, darwin, win32).

Who created Engineering?

It is built and maintained by Iván (@ivangdavila); the current version is v1.0.0.

💬 Comments