CtrlK
BlogDocsLog inGet started
Tessl Logo

phoenix-cli

Debug LLM applications using the Phoenix CLI. Fetch traces, analyze errors, structure trace review with open coding and axial coding, inspect datasets, review experiments, query annotation configs, and use the GraphQL API. Use whenever the user is analyzing traces or spans, investigating LLM/agent failures, deciding what to do after instrumenting an app, building failure taxonomies, choosing what evals to write, or asking "what's going wrong", "what kinds of mistakes", or "where do I focus" — even without naming a technique.

70

Quality

Does it follow best practices?

Impact

No eval scenarios have been run

SecuritybySnyk

Passed

No known issues

SKILL.md
Quality
Evals
Security

Quality

Content

72%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is highly actionable with executable commands and real reference-file disclosure, but loses points on conciseness from mild over-explanation and on workflow clarity because the destructive annotation-delete path lacks an explicit validation checkpoint.

Suggestions

Trim explanatory prose that restates what Claude already knows (e.g. the profiles intro paragraph) to lean command-plus-shape reference material.

Add an explicit validation step before the destructive `*-annotations delete --all` operations — e.g. a dry-run/list to confirm the identifier matches the expected artifacts before running the three DELETEs.

Consolidate the repeated `--identifier "<coding-annotation-id>"` tagging comments into one stated convention to reduce repeated inline explanation.

DimensionReasoningScore

Conciseness

The body is mostly efficient command tables and JSON shapes, but it re-explains some concepts Claude already knows (e.g. "Named profiles let you switch between multiple Phoenix instances without juggling environment variables") and could be tightened in the prose around profiles and annotation configs.

2 / 3

Actionability

Nearly every section gives copy-paste-ready, executable `px` commands with concrete flags and jq pipelines, plus exact JSON shapes — fully executable guidance rather than abstract description.

3 / 3

Workflow Clarity

The open-coding → axial-coding → evals workflow is sequenced, but the destructive batch DELETE operations ("nuke every annotation tied to this coding annotation identifier") rely on a single opt-in confirmation note rather than an explicit validate-then-act checkpoint with error recovery, capping clarity at 2.

2 / 3

Progressive Disclosure

SKILL.md is an overview with two clearly signaled, one-level-deep references (references/open-coding.md, references/axial-coding.md), both confirmed to exist, and detailed coding methodology is appropriately split out rather than inlined.

3 / 3

Total

10

/

12

Passed

Description

100%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is specific, trigger-rich, and explicitly pairs capabilities with a 'Use whenever' clause, hitting the top anchor on every dimension. It uses third person and avoids vague fluff or over-claims.

DimensionReasoningScore

Specificity

Lists multiple concrete actions — "Fetch traces, analyze errors, structure trace review with open coding and axial coding, inspect datasets, review experiments, query annotation configs, and use the GraphQL API" — matching the anchor for listing several specific concrete actions.

3 / 3

Completeness

Clearly answers both what it does (the enumerated CLI actions) and when to use it via an explicit "Use whenever..." trigger clause, satisfying the both-what-and-when anchor.

3 / 3

Trigger Term Quality

Covers natural phrases users would say — "analyzing traces or spans", "investigating LLM/agent failures", "what's going wrong", "what kinds of mistakes", "where do I focus" — giving good coverage of natural terms beyond technical jargon.

3 / 3

Distinctiveness Conflict Risk

The Phoenix CLI / LLM-trace-analysis niche with distinct triggers (instrumenting an app, building failure taxonomies, choosing evals) is unlikely to collide with other skills.

3 / 3

Total

12

/

12

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

referenced_paths_exist

Referenced path issues: 4 missing

Warning

Total

15

/

16

Passed

Repository
Arize-ai/phoenix
Reviewed

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.