agentsclimarketplace
Sign in

Detect performance signals

Skill alexe-ev/product-plugins/data-analytics/skills/detect-performance-signals

Identify and interpret signals of product performance change including regressions, improvements, and anomalies. Use this skill when a team needs to investigate an unexpected change in a metric.From its SKILL.md

Install
npx -y skills add alexe-ev/product-plugins --skill detect-performance-signals

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.6 KB, 480 tokens by cl100k_base, as published. Nobody here has run it

Detect Performance Signals

Purpose

Help teams identify, contextualize, and interpret signals of performance change — distinguishing real shifts from noise, and regressions from improvements.

Skill type

Conceptual skill with calculation-aware components

Use this skill when

  • A metric has moved unexpectedly (up or down)
  • A team is unsure whether a change is meaningful or noise
  • An alert has fired and needs to be investigated
  • Post-launch monitoring has surfaced an anomaly

Do not use this skill when

  • The goal is structured experiment result analysis (use analyze-experiment-results)
  • No metric baseline or monitoring exists

Required inputs

  • Metric that changed
  • Observed change (magnitude and direction)
  • Time period of the change

Optional inputs

  • Baseline or expected value
  • Recent product changes or events
  • Segment breakdown
  • Data source and methodology

Upstream context

Works best when:

  • Metrics framework is defined
  • Dashboard is in place
  • Historical baseline exists

If upstream context is missing

If no baseline exists, produce a signal investigation framework and flag that without baselines, changes cannot be assessed.

Downstream handoff

Output can feed:

  • formulate-experiment-hypothesis (unexplained signal → investigation → experiment)
  • analyze-funnel-retention-cohorts (drill into funnel or cohort to diagnose)
  • identify-problem-opportunity (signal becomes a product problem to investigate)

Instructions

  1. Establish the baseline (what was normal before the change).
  2. Quantify the change: magnitude, direction, duration.
  3. Check for data quality issues: tracking change, instrumentation error, data lag.
  4. Identify potential causes: product changes, external events, seasonality, data issues.
  5. Determine if the change is statistically meaningful vs. noise.
  6. Recommend investigation path: more data, qualitative research, or experiment.

Output

Provide:

  • Signal summary: what changed, by how much, when
  • Baseline context
  • Data quality assessment
  • Potential causes (ordered by likelihood)
  • Assessment: noise / meaningful signal / regression / improvement
  • Recommended next steps

Risks / caveats

  • Avoid premature conclusions — investigate causes before acting
  • Check instrumentation before assuming a real product change
  • Seasonality and external events are common and frequently overlooked

What ships with it: 5 files

11.9 KB alongside SKILL.md

examples/

Keep looking

Skills are one crate of 326,861. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.