SKILLEMALL.ai

AB testing-in-production

Safe-release techniques DURING rollout: feature flags, progressive rollouts, canary analysis, guardrail metrics, production smoke tests, and synthetic users. Bridges QA and SRE practices. Use when: "feature flag testing," "canary deploy," "progressive rollout," "guardrail metrics," "dark launch," "safe rollout." Not for: scheduled probes that run continuously after release — use `synthetic-monitoring`. Not for: designing tests from prod telemetry — use `observability-driven-testing`. Related: release-readiness, synthetic-monitoring, observability-driven-testing, qa-metrics.

petrkindlmann/qa-skills Agent Skills author: petrkindlmann MIT 3 files body ≈ 5 086 tokens Open the sourcegithub.com analyzed 2 d ago

Safe-release techniques DURING rollout: feature flags, progressive rollouts, canary analysis, guardrail metrics, production smoke tests, and synthetic users.

As a process B 74/100 · Nearly there — weak spots: result and completion, progress reporting

AnalyzerInfrastructureData and analyticstype and topics are labelled automatically from the skill text
JSON
Technical rating
A
95/100
safety, quality, tests
Safety 60%
100
Quality 40%
88
Run on models
none yet
Process rating
B
74/100
Nearly there
Progress reporting w 2
0
Result and completion w 14
40
When it triggers w 12
50
the three weakest of ten parameters · all ten

How to improve

  1. The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
For the model run — optional
  • Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
  • A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.

Guard findings · 0

✓ No critical or high findings

Files scanned: 3. Evidence is masked. Grey chips explain why severity was lowered.

Against the Agent Skills spec

  • warning body-long SKILL.md body ≈ 5086 tokens (recommended < 5000); move details to references/
  • note edit-residue the text marks something as outdated (lines 103): check that old rules are not kept next to new ones — the full check reads the text for contradictions

Process rating: all ten parameters 74/100

  • 0Progress reporting. Says nothing while it works
  • 40Result and completion. Does not say what the result is
  • 50When it triggers. No condition that starts the skill
  • 50Failures and branches. 0 branches, has a failure section
  • 70Inputs and preconditions. Inputs and preconditions are listed
  • 70Execution cost. Instruction body is 5086 tokens
  • 100Tools and files. No external tools needed
  • 100Steps. 38 steps
  • 100Consistency. Name and required fields are in place
  • 100Running it twice. Mutating operations check current state
  • medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
  • low 13 top-level sections: this looks like several domains in one skill

Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.

Quality signals

  • +3Output format is not stated: the model decides each time
  • +2Single-language instructions
  • +5Description quotes 6 example trigger phrases
  • +4Description says when NOT to use the skill
  • +3Description length 580: enough signal without eating the budget
  • +4Structure: 40 headings
  • +3Step-by-step instructions: 38 items
  • +4Has examples (5 code blocks)
  • +4Reference files are cited in the instructions (2 of 2)
  • +1License stated

Quality base 70; lint remarks subtract, signals add up to 100. Result: 88.