RBraga01

Assign AP table ratings (H-AP, M-AP, L-AP) in PFMEA or DFMEA — the RPN replacement from AIAG-VDA FMEA Handbook 2019. Explains H/M/L classification logic, mandatory action requirements for High Priority items, and OEM-specific thresholds. Use when assigning risk levels in PFMEA or DFMEA, or auditing FMEA documents for AP compliance.

Run quality problem zeroing, perform 双归零 (double-five zeroing), write a technical zeroing report, write a management zeroing report, or close an aerospace quality problem under GB/T 29076—2021. Covers problem reporting, emergency response, team formation, technical zeroing (定位准确、机理清楚、问题复现、 措施有效、举一反三), management zeroing (过程清楚、责任明确、措施落实、严肃处理、 完善规章), evidence gates, review, and closure. Use for aerospace or complex equipment failures, defects, accidents, repeated problems, human-responsibility problems, and issues requiring a formal closed-loop investigation.

Interactive Control Plan builder — takes PFMEA failure modes and process flow steps as input and builds a complete Control Plan row by row, with correct control methods, sample plans, and reaction plans for each characteristic. Use when building a new Control Plan, updating after a corrective action (8D D7), or converting a PFMEA into Control Plan format. Covers AIAG Control Plan reference manual and IATF 16949 §8.5.1.

Interactive PPAP completeness checker — walks through all 18 PPAP elements for a given submission level, identifies missing or incomplete items, and generates a gap report before PSW signature. Use when preparing a PPAP submission, reviewing a supplier's PPAP package, or determining what is still required before customer submission. Covers AIAG PPAP 4th Edition.

VDA 6.3 Process Audit — conduct or prepare for a process audit using the VDA 6.3 methodology, evaluate process elements P1–P7, calculate degree of fulfillment, classify findings, and generate an audit report. Use when a customer requests a VDA 6.3 audit, when auditing a supplier's manufacturing process, or when preparing for a VDA process audit visit. Covers VDA 6.3 4th edition (2023).

Measurement System Analysis (MSA) and Gauge Repeatability & Reproducibility (Gauge R&R) — plan, execute, and interpret an MSA study for variable or attribute measurement systems. Use when qualifying a gauge for a new part, validating a measurement system before PPAP, interpreting Gauge R&R results, or auditing MSA studies for adequacy. Covers AIAG MSA 4th edition.

Statistical Process Control (SPC) — select the correct control chart, interpret out-of-control signals using Western Electric rules, calculate and interpret Cp, Cpk, Pp, Ppk. Use when setting up SPC for a new characteristic, interpreting control chart signals, responding to special cause variation, or auditing SPC implementation. Covers AIAG SPC 2nd edition and IATF 16949 §8.3.3.

Advanced Product Quality Planning (APQP) — plan and track a new product launch through all 5 phases, identify deliverables per phase, run gate reviews, and ensure quality outputs are complete before Start of Production (SOP). Use when launching a new part, managing an APQP project, or auditing APQP completeness. Covers AIAG APQP 2nd edition and IATF 16949 §8.3.

Systematic architecture review workflow. Use before major feature work on an unfamiliar or inherited codebase, before scaling milestones, or as a periodic health check. Maps the system as-built (not as-documented), dispatches parallel specialist reviews, and produces an evidence-backed findings report with severity verdicts.

The meta-skill. Loaded at every session start. Defines which A Team skills and agents MUST be used for which situations. If a skill or agent applies, you do not have a choice — you must use it.

Conversational onboarding for A Team. Detects ROADMAP.md, extracts context, generates INIT.md without requiring technical knowledge. Invoked automatically by the orchestrator when INIT.md is missing.

Root cause analysis using the Five Whys technique. Use when a bug persists despite surface fixes, a failure recurs, or a process keeps breaking. Ensures the fix targets the root, not the symptom.

Hard gate before implementing any API endpoint or service interface. Requires a written, reviewed contract (OpenAPI 3.x, protobuf, or GraphQL schema) before any implementation code is written. Prevents breaking changes, misaligned clients, and undocumented behaviour.

Use BEFORE any creative work — creating features, building components, adding functionality. Explores user intent, requirements, and design before implementation. Hard-gated: no code until user approves the spec.

Hard gate before any database schema change or data transformation in production. Requires a written migration plan with rollback strategy, dry run, and verification steps before execution. Prevents data loss, downtime, and irreversible schema changes.

Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies. Dispatches one focused agent per problem domain, runs them concurrently, then integrates.

Use before launching any LLM feature or when monthly API costs are growing unexpectedly. Requires token count measurement, call volume analysis, and cost projection at 10× scale. Blocks "it's cheap enough now" completions.

Use before shipping any LLM feature that touches users. Reviews prompt injection, hallucination risk, output misuse, agentic scope, and abuse vectors. Blocks "nobody will try that" completions.

Use when prompt cost is too high, latency is above threshold, or context window limits are being approached. Requires measurement before and after each reduction. Blocks "I shortened the prompt so it should be cheaper" completions.

Use before merging, deploying, or demo'ing any LLM feature. Requires documented eval results — pass rate, failure analysis, baseline comparison. Blocks "it looked good when I tested it" completions.

Use before merging any PR that adds an LLM API call. Every call must handle timeout, malformed output, low confidence, and refusal — with a defined, user-safe fallback for each. Blocks "add error handling later" completions.

Use when selecting a model for any production feature, or evaluating whether to switch models. Requires task-specific benchmarking — not leaderboard lookup. Blocks "GPT-4 is the best model" decisions.

Use whenever writing or modifying a prompt that will run in production. Enforces version-controlled prompts in prompts/<feature>/v<x.y.z>.md. Blocks "the prompt is in the code somewhere" completions.

Use when designing or auditing a retrieval-augmented generation pipeline. Requires data audit and query audit before any design decision. Blocks "I'll use the standard setup" completions.