CtrlK
BlogDocsLog inGet started
Tessl Logo

phoenix-client-development

Guide for the phoenix-client TypeScript package — experiment lifecycle, tracer provider management, and test conventions.

68

1.23x
Quality

52%

Does it follow best practices?

Impact

100%

1.23x

Average score across 3 eval scenarios

SecuritybySnyk

Passed

No known issues

Optimize this skill with Tessl

npx tessl skill review --optimize ./js/packages/phoenix-client/.agents/skills/phoenix-client-development/SKILL.md
SKILL.md
Quality
Evals
Security

Evaluation results

100%

17%

Experiment Orchestrator Implementation

Two-phase experiment provider lifecycle

Criteria
Without context
With context

Two phases exist

100%

100%

Task projectName

100%

100%

Eval projectName

0%

100%

global: false on providers

100%

100%

No direct register() global

100%

100%

Task cleanup before eval start

100%

100%

Cleanup order: flush then shutdown

30%

100%

detach in finally

100%

100%

Creator owns cleanup

100%

100%

Notes explain lifecycle

100%

100%

100%

6%

Unit Test Suite for Experiment Runner

Unit test mock shapes and tracing assertions

Criteria
Without context
With context

Test file location

100%

100%

register() mock shape

40%

100%

attachGlobalTracerProvider mock shape

100%

100%

forceFlush returns promise

100%

100%

shutdown returns promise

100%

100%

register() called twice

100%

100%

register() called with global: false

100%

100%

Task phase projectName

100%

100%

Eval phase projectName

100%

100%

detach called once per handle

100%

100%

attachGlobalTracerProvider per phase

100%

100%

100%

32%

Integration Tests for Provider Lifecycle Restoration

Integration test conventions and sentinel provider pattern

Criteria
Without context
With context

integration- prefix

0%

100%

No vitest imports

100%

100%

Plain assert helper

100%

100%

No console.log

100%

100%

process.exit(1) on failure

100%

100%

trace.disable() cleanup

0%

100%

context.disable() cleanup

0%

100%

propagation.disable() cleanup

0%

100%

Sentinel provider present

100%

100%

Integration notes

100%

100%

Repository
Arize-ai/phoenix
Evaluated
Agent
Claude Code
Model
Claude Sonnet 4.6

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.