SkillVaultskills Browse all 500 skills

Architecture · Version 1.2.0 · Reviewed 2026-08-02

Read Model Projection Advisor

Make a defensible decision about projection design and rebuild strategy with evidence, explicit trade-offs, and a verification plan.

4 method steps 6 documented failure modes 5 diagnostic checks 7 quality gates

Designs denormalized read models with rebuild strategy, staleness bounds, and consistency signals.

₹99 one-time

Get this skill archive

What this skill helps you do

  • Projection design
  • Rebuild strategy
  • Staleness bounds

How Read Model Projection Advisor works

You provide

Prompts, model versions, evaluation data, and failures

It inspects

Failure class and context sufficiency for projection design

It decides

A rebuild strategy change with one variable moved

You verify

Pass rate per case class against a pinned baseline

What it checks first

Read Model Projection Advisor designs denormalized read models with rebuild strategy, staleness bounds, and consistency signals. Use it when the work involves Projection design, Rebuild strategy, Staleness bounds.

  1. Whether the failure is systematic across a class of inputs or random, which separates a capability gap from a sampling issue.
  2. Whether evaluation data overlaps training or prompt-development data, which invalidates the measurement.
  3. Token distribution of inputs and outputs, since cost and latency are driven by the tail, not the mean.
  4. Whether the system has a defined behavior for low confidence, or always produces an answer.
  5. Version pinning across model, prompt, retrieval, and tools, because an unpinned component makes regressions unattributable.

Failure modes it recognizes

  • Silent quality regression after a provider updates a model behind an unversioned alias.
  • Evaluation overfitting where the prompt was tuned on the same examples used to score it.
  • Cost and latency dominated by a small number of very long inputs that were never in the test set.
  • Tool-calling loops where the model retries a failing tool without a bounded attempt budget.
  • Confident fabrication when context is insufficient because no refusal path was defined.
  • Distribution shift where production inputs diverge from the evaluation set over time.

Answers it will reject

  • Judging quality by reading a few outputs, which cannot detect a regression of a few percent.
  • Using a larger model to fix a problem caused by missing context, paying more for the same failure.
  • Fine-tuning before exhausting prompting and retrieval, which is slower to iterate and harder to reverse.
  • Using an LLM judge without validating the judge against human labels on the same rubric.

Decision rules it applies

  • Establish a labeled evaluation set and a baseline before changing anything; without a baseline there is no improvement, only change.
  • Pin every version and change one component at a time.
  • Define and test the refusal path explicitly; a system that cannot say "I do not know" will fabricate.
  • Budget latency and cost on p95 token counts, not averages.

Evidence it asks for

  • Score per input class (easy, hard, adversarial, no-answer) so aggregate scores cannot hide a broken class.
  • Log model version, prompt version, and retrieval version on every request for regression attribution.
  • Track p50 and p95 tokens and cost per successful task, not per call.

The method inside

  1. Map the artifact, actors, boundaries, and invariants relevant to projection design.
  2. Trace concrete failure or abuse paths for rebuild strategy; do not report checklist items without a mechanism.
  3. Prioritize staleness bounds findings by impact, likelihood, confidence, and cost of correction.
  4. Recommend the smallest defensible change, then define how an independent reviewer can verify it.

Deliverables

  • Projection design assessment
  • Rebuild strategy decision and action plan
  • Staleness bounds verification checklist

Evidence requirements

  • Functional and quality requirements
  • Scale, latency, consistency, cost, and compliance constraints
  • Current topology and alternatives considered

Quality gates

  • Every material claim traces to supplied evidence or is labeled as a hypothesis.
  • The response follows the declared deliverable contract.
  • No execution, access, measurement, or verification is invented.
  • Secrets and personal data are redacted rather than repeated.
  • The user receives a concrete independent verification step.
  • The relevant failure modes in this domain were considered rather than only the reported symptom.
  • No listed anti-pattern was recommended as a solution.

Example task

Input

Our read model drifts from the source and rebuilding it means several hours of stale results.

Expected output

Drift plus painful rebuild means projections are updated in place with no replay path. Build projections as a pure function of an ordered event log so a rebuild is deterministic, and rebuild into a new version behind an alias to avoid downtime...

Boundaries and compatibility

Ideal for

  • Projection design: produce a decision or artifact grounded in supplied evidence.
  • Rebuild strategy: produce a decision or artifact grounded in supplied evidence.
  • Staleness bounds: produce a decision or artifact grounded in supplied evidence.

Out of scope

  • Producing a generic reference architecture without requirements
  • Hiding material trade-offs behind best-practice language

Agent compatibility

  • GitHub Copilot custom agents
  • Claude Agent Skills / SKILL.md
  • Any instruction-following chat model

Tool policy: Advisory by default. No tools are assumed. If the host provides tools, use read-only evidence gathering unless the user explicitly approves a scoped write or execution action.