メニュー
Template skill — copy this folder to start a new quality-platform skill. Demonstrates scoring one FMEA Severity/Occurrence/Detection triple by calling the fmea_score MCP tool; do not install directly.
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
直接コマンドでは確認用 Prompt が省略されます。実行前にソースを確認してください。
npx skills add https://github.com/Siddardth7/quality-platform --skill example-skillコマンドは1行のまま表示されます。コピー前に横へスクロールして全体を確認してください。
ローカルで確認しますか？SkillsMP が現在取得できるファイルをダウンロードできます。
SKILL.md を表示中
Answer a quality-standards question (AIAG FMEA/SPC/MSA thresholds, ISO/NIST concepts, why a rule like the ndc threshold of 5 or an RPN/%GRR band exists) by calling the quality-platform qdb_answer_question MCP tool over HTTP for a cited, page-located answer; falls back to this project's own public ASSUMPTIONS_LOG citations when the endpoint is unavailable. Use for "why" and "what does the standard say" questions, not for running fmea_score, an SPC chart, msa_gage_rr or controlplan_build.
Score or rank Failure Mode and Effects Analysis (FMEA) rows by calling the quality-platform FMEA MCP tools — Risk Priority Number (RPN) and AIAG-VDA Action Priority (AP) for one Severity/Occurrence/Detection triple, a full DFMEA/PFMEA table, or a relational failure-mode model. Use for failure mode, root cause and risk-prioritization requests, not statistical process control or Gage R&R.
Validate a measurement system by calling the quality-platform MSA MCP tool — crossed Gage R&R via Average-and-Range or ANOVA, returning %GRR (study-variation and tolerance bases), ndc, EV/AV/PV components and an AIAG-cited Accept/Marginal/Reject verdict. Use for gauge study, repeatability & reproducibility, %GRR/ndc and measurement-system-capability requests, not control charts/process capability or FMEA risk scoring.
| name | example-skill |
| description | Template skill — copy this folder to start a new quality-platform skill. Demonstrates scoring one FMEA Severity/Occurrence/Detection triple by calling the fmea_score MCP tool; do not install directly. |
This is the copy-from template for quality-platform skills. It shows the whole pattern in the
smallest possible case: the skill orchestrates, the engine decides. Read
skills/CONVENTIONS.md before you edit a copy of this folder.
The user supplies one Severity / Occurrence / Detection triple and wants its Risk Priority
Number and AIAG-VDA Action Priority. For a whole FMEA table, use the fmea_run tool instead.
python -m mcp_app.server from the workspace root.fmea_list_scales,
fmea_get_scale), not in this skill.fmea_score tool with severity, occurrence and detection.rpn and action_priority verbatim. Never recompute, round,
re-derive or "sanity check" either value — the engine owns that arithmetic, and a second
opinion computed here is a defect, not a safeguard.scripts/call_fmea_score.py is the same flow as a runnable script, for when a shell call is
cheaper than a tool call.
references/mcp-tool-contract.md — the tool namespace
convention, the fmea_score request/response shape, and the error contract.