SkillVaultskills Browse all 500 skills

Reliability · Version 1.6.0 · Reviewed 2026-08-02

Crash-Free Rate Advisor

Reduce production risk in cluster ranking and user impact with evidence, explicit trade-offs, and a verification plan.

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

Improves mobile stability by ranking crash clusters against affected users rather than event counts.

₹99 one-time

Get this skill archive

What this skill helps you do

  • Cluster ranking
  • User impact
  • Release gating

How Crash-Free Rate Advisor works

You provide

Impact window, telemetry, and dependency state

It inspects

Saturation and blast radius behind cluster ranking

It decides

A user impact plan that stabilizes before diagnosing

You verify

Detect, mitigate, and resolve times recorded separately

What it checks first

Crash-Free Rate Advisor improves mobile stability by ranking crash clusters against affected users rather than event counts. Use it when the work involves Cluster ranking, User impact, Release gating.

  1. User-visible impact and error-budget consumption rather than component health.
  2. Saturation signals — queue depth, pool utilization, connection counts — near the onset.
  3. Whether the system recovered on its own, which indicates saturation rather than corruption.
  4. The blast radius and what boundary should have contained it.

Failure modes it recognizes

  • Retry amplification turning a partial failure into a total outage.
  • A shared dependency creating correlated failure across supposedly independent services.
  • Slow resource exhaustion invisible until a hard limit is crossed.
  • A rollback blocked by an incompatible migration.

Answers it will reject

  • Treating the trigger as the root cause, which stops the analysis before the fragility is identified.
  • Adding a runbook step where a boundary would remove the failure mode.
  • Measuring availability as a mean, which hides regional and tenant-level outages.

Decision rules it applies

  • Stabilize user impact before completing diagnosis.
  • Bound every retry with a budget, jitter, and a circuit breaker.
  • Prefer removing a failure mode over detecting it faster.

Evidence it asks for

  • Record time-to-detect, time-to-mitigate, and time-to-resolve separately.
  • Quantify impact in customer terms: failed requests, affected accounts, duration.
  • Verify recovery with the same signal that detected the failure.

The method inside

  1. Map the artifact, actors, boundaries, and invariants relevant to cluster ranking.
  2. Trace concrete failure or abuse paths for user impact; do not report checklist items without a mechanism.
  3. Prioritize release gating 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

  • Cluster ranking assessment
  • User impact decision and action plan
  • Release gating verification checklist

Evidence requirements

  • User-visible symptoms and SLO impact
  • Timeline, telemetry, deploys, and dependency state
  • Current mitigations and operational constraints

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 crash-free rate is 98.5 percent and we have hundreds of distinct crash signatures to work through.

Expected output

Event counts over-weight crash loops from a few devices, so rank by affected users instead. That usually collapses hundreds of signatures into a handful worth fixing, and gives you a release gate you can defend...

Boundaries and compatibility

Ideal for

  • Cluster ranking: produce a decision or artifact grounded in supplied evidence.
  • User impact: produce a decision or artifact grounded in supplied evidence.
  • Release gating: produce a decision or artifact grounded in supplied evidence.

Out of scope

  • Replacing incident command authority
  • Calling a trigger the root cause without a causal chain

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.