Skip to content
Back to skills

Mmpsd Allowables

ASecurity

Use when computing statistically based metallic material design allowables per MMPDS: determine A-basis and B-basis values from coupon test data, run the one-sided normal tolerance k-factor approximation, and validate that derived allowables sit below the sample mean. The skill applies minimum sample counts, confidence/content conventions, and sanity checks on design values, following the statistical approach of MMPDS and its MIL-HDBK-5 heritage. Trigger: allowables, a-basis, b-basis, k-facto...

  • 2 stars
  • 0 votes
  • 0 copies
  • 3 views
  • Added September 27, 2026
ai-agentspython

Works with

  • claude code

Security analysis

A100/100

Pro scans all 3 files and shows the line behind each finding

Scanned September 27, 2026

npx -y skills add ashfordeOU/aero-agent-skills --skill mmpsd-allowables --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Mmpsd Allowables?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Mmpsd Allowables
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/ashfordeou-mmpsd-allowables/badge)](https://www.skillsdirectory.com/skills/ashfordeou-mmpsd-allowables)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: mmpsd-allowables
description: "Use when computing statistically based metallic material design allowables per MMPDS: determine A-basis and B-basis values from coupon test data, run the one-sided normal tolerance k-factor approximation, and validate that derived allowables sit below the sample mean. The skill applies minimum sample counts, confidence/content conventions, and sanity checks on design values, following the statistical approach of MMPDS and its MIL-HDBK-5 heritage. Trigger: allowables, a-basis, b-basis, k-factor, metallic materials, sample statistics, design values, mil-hdbk-5, test specimens."
license: Apache-2.0
compliance: STANDARDS-REF
standards:
  - id: mmpsd
    reference-only: true
gated: false
domain: structures
pack: structures
compatibility: "agentskills.io SKILL.md; any SKILL.md host (Claude Code, Hermes, OpenClaw)"
metadata:
  domain: structures
  subdomain: materials
  tags: [allowables, a-basis, b-basis, k-factor, metallic-materials, sample-statistics, design-values, mil-hdbk-5, test-specimens]
  version: 0.1.0
  author: Aero Agent Skills
---

# MMPDS Metallic Allowables (structures/materials/mmpsd-allowables)

Use when the task is statistically based metallic material design
allowables: A-basis and B-basis values from coupon test samples,
one-sided normal tolerance k-factors, and design-value sanity
checks.

## Domain quick reference

- A-basis: 95% confidence that 99% of the population exceeds the
  value; B-basis: 95% confidence that 90% exceeds the value.
- One-sided normal tolerance k-factor (Owen/Odeh approximation):
  z_c = 1.6448536269514722 (0.95 confidence); z_p = 2.3263478740408408
  for A-basis content or 1.2815515655446004 for B-basis;
  a = 1 - z_c^2/(2(n-1)); b = z_p^2 - z_c^2/n;
  k = (z_p + sqrt(z_p^2 - a*b))/a.
- Allowable = sample mean - k * sample standard deviation.
- Common minimum samples: 10 for A-basis, 6 for B-basis; verify
  against the current MMPDS edition.
- MMPDS is the successor to MIL-HDBK-5 for metallic allowables.

## Workflow

1. Gather the coupon test sample for the material, condition, and
   property of interest; check the sample count against
   min_samples.
2. Compute the k-factor with k_factor_one_sided for the basis.
3. Compute the allowable with allowable_from_sample.
4. Sanity-check the result with design_value_sanity: the allowable
   must be positive and below the sample mean.
5. Confirm the deterministic checks with the contract test
   scripts/test_mmpsd.py.

## Pitfalls

- Using a B-basis k-factor when A-basis is required (A is always
  lower/more conservative).
- Computing k-factors from samples below the minimum count.
- Applying the normal tolerance method to skewed or multimodal
  data without transformation.
- Copying design-value tables from MMPDS; derive values from the
  sample statistics instead.

## Behavior contract (gate 3)

The k-factor, sample, and sanity-check logic is exercised by the
gate 3 contract test: scripts/test_mmpsd.py against
scripts/mmpsd_logic.py (stdlib unittest, offline). Run:
python3 scripts/test_mmpsd.py

## Compliance

- Standards referenced, not reproduced: MMPDS is proprietary
  (SAE, successor to public-domain MIL-HDBK-5); name + paraphrase
  + link only per standards-map.yaml and brief 06; never
  reproduce design-value tables.
- compliance: STANDARDS-REF, gated: false.

Files in this skill

  • SKILL.md3.3 KB
  • scripts/mmpsd_logic.py2.8 KB
  • scripts/test_mmpsd.py3.8 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

View all in ai-agents →

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.